Cardiometabolic medication exposures and cognitive outcomes in Alzheimer's disease.

Code ↔ Paper

5 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 5 matches

[1] § Methods › Statistical models ↔ code/Forest_plots_CDR(cognitive).Rmd, lines 498–552 · score 0.75 · Dem progressors, ADem, clinical progression, MCI progressors, parallel, tracks
[2] § Methods › Study design ↔ code/preprocessed_HABSHD.ipynb, lines 39–48 · score 0.75 · thyroid disease, high cholesterol, anxiety, depression, stroke, hypertension
[3] § Methods › Study design ↔ code/preprocessed_NACC.ipynb, lines 50–59 · score 0.67 · high cholesterol, anxiety, depression, stroke, thyroid, hypertension
[4] § Methods › Statistical models ↔ code/Forest_plots_CDR(cognitive).Rmd, lines 498–552 · score 0.60 · ADem, clinical progression, MCI progression, stratified, backgrounds, Cognitive
[5] § Methods › Statistical models ↔ code/Forest_plots_CDR(baseline cognitive).Rmd, lines 317–369 · score 0.52 · ADem, parallel, tracks, Diamond, blocks, forest

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The paper is loaded by your browser, with JavaScript: read it at doi.org or on Europe PMC.

The authors' code

R Markdown · 553 lines · 19 KB · no license · 2 matches

---
title: "Forest Plots - Cognitive Progression Groups"
author: "Yihan Wang"
date: "2025-09-16"
output:
word_document: default
pdf_document: default
---
1. PROGRESSION GROUPS (4 groups):
Groups included:
- Non-progression (CU stable + MCI stable combined)
- CU-MCI progression (progression from CU to MCI)
- CU/MCI-AD progression (progression to AD dementia)
- AD (dementia patients)
2. COMORBIDITY ADJUSTMENTS:
Added to ALL models as covariates:
- CVD (Cardiovascular disease)
- Endocrine (Endocrine disorders including diabetes)
- Psychiatric (Psychiatric conditions)
3. MEDICATION × TIME INTERACTIONS:
All 7 medications include interaction with year_since_baseline:
- ACEi × Time, ARB × Time, β-Blocker × Time, CCB × Time
- Diuretic × Time, Metformin × Time, Statin × Time
This allows assessment of whether medication effects change over follow-up period.
4. MODEL FORMULA STRUCTURE:
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
drug:year_since_baseline (for all 7 drugs) +
(1 + year_since_baseline | id)
```{r}
rm(list=ls())
library(missRanger)
library(dplyr)
library(lubridate)
library(lme4) # mixed models
library(lmerTest) # p-values for lmer
library(splines)
library(conflicted)
library(broom.mixed)
library(dplyr)
library(stringr)
library(forcats)
library(ggplot2)
library(forestploter)
library(grid)
library(gridtext)
conflicts_prefer(dplyr::lag)
conflicts_prefer(lmerTest::lmer)
conflicts_prefer(dplyr::filter)
```
```{r}
source("medication-lmer-utility.R")
source("plots.R")
```
```{r}
NACC_full_df_cdr<-read.csv("../preprocessed_data/NACC/NACC_cdr_imputed_binary.csv")
AIBL_full_df_cdr<-read.csv("../preprocessed_data/AIBL/AIBL_cdr_imputed_binary.csv")
HABS_full_df_cdr<-read.csv("../preprocessed_data/HABSHD/HABSHD_cdr_imputed_binary.csv")
NACC_full_df_cdr$edu <- scale(NACC_full_df_cdr$edu)
NACC_full_df_cdr$age <- scale(NACC_full_df_cdr$age)
AIBL_full_df_cdr$edu <- scale(AIBL_full_df_cdr$edu)
AIBL_full_df_cdr$age <- scale(AIBL_full_df_cdr$age)
HABS_full_df_cdr$edu <- scale(HABS_full_df_cdr$edu)
HABS_full_df_cdr$age <- scale(HABS_full_df_cdr$age)
```
```{r}
# Split by progression groups
NACC_split <- split_by_progression(NACC_full_df_cdr, id_col = "id", visit_col = "visit_no", status_col = "status")
AIBL_split <- split_by_progression(AIBL_full_df_cdr, id_col = "id", visit_col = "visit_no", status_col = "status")
HABS_split <- split_by_progression(HABS_full_df_cdr, id_col = "id", visit_col = "visit_no", status_col = "status")
# Extract groups (function now directly returns the 4 groups we need)
# Non-progression (stable CU + stable MCI combined by the function)
NACC_non_prog = NACC_split$groups$Non_progression
AIBL_non_prog = AIBL_split$groups$Non_progression
HABS_non_prog = HABS_split$groups$Non_progression
# CU-MCI progression
NACC_cu_mci_prog = NACC_split$groups$CU_MCI_progression
AIBL_cu_mci_prog = AIBL_split$groups$CU_MCI_progression
HABS_cu_mci_prog = HABS_split$groups$CU_MCI_progression
# CU/MCI-AD progression
NACC_to_ad_prog = NACC_split$groups$CU_MCI_AD_progression
AIBL_to_ad_prog = AIBL_split$groups$CU_MCI_AD_progression
HABS_to_ad_prog = HABS_split$groups$CU_MCI_AD_progression
# AD
NACC_ad = NACC_split$groups$AD_stable
AIBL_ad = AIBL_split$groups$AD_stable
HABS_ad = HABS_split$groups$AD_stable
```
# Full
```{r}
NACC_full_df_cdr$visit_date <- as.Date(NACC_full_df_cdr$visit_date, format = "%Y-%m-%d")
fit_full_NACC <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = NACC_full_df_cdr
)
summary(fit_full_NACC)
```
```{r}
AIBL_full_df_cdr$visit_date <- as.Date(AIBL_full_df_cdr$visit_date, format = "%Y-%m-%d")
fit_full_AIBL <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = AIBL_full_df_cdr
)
summary(fit_full_AIBL)
```
```{r}
HABS_full_df_cdr$visit_date <- as.Date(HABS_full_df_cdr$visit_date, format = "%Y-%m-%d")
fit_full_HABS <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = HABS_full_df_cdr
)
summary(fit_full_HABS)
```
# Non-progression (CU + MCI stable combined)
```{r}
NACC_non_prog$visit_date <- as.Date(NACC_non_prog$visit_date, format = "%Y-%m-%d")
fit_NACC_NON_PROG <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = NACC_non_prog
)
summary(fit_NACC_NON_PROG)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_NACC_NON_PROG))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
```{r}
AIBL_non_prog$visit_date <- as.Date(AIBL_non_prog$visit_date, format = "%Y-%m-%d")
fit_AIBL_NON_PROG <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = AIBL_non_prog
)
summary(fit_AIBL_NON_PROG)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_AIBL_NON_PROG))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
```{r}
HABS_non_prog$visit_date <- as.Date(HABS_non_prog$visit_date, format = "%Y-%m-%d")
fit_HABS_NON_PROG <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = HABS_non_prog
)
summary(fit_HABS_NON_PROG)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_HABS_NON_PROG))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
# CU-MCI Progression
```{r}
NACC_cu_mci_prog$visit_date <- as.Date(NACC_cu_mci_prog$visit_date, format = "%Y-%m-%d")
fit_NACC_CU_MCI_PROG <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = NACC_cu_mci_prog
)
summary(fit_NACC_CU_MCI_PROG)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_NACC_CU_MCI_PROG))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
```{r}
AIBL_cu_mci_prog$visit_date <- as.Date(AIBL_cu_mci_prog$visit_date, format = "%Y-%m-%d")
fit_AIBL_CU_MCI_PROG <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = AIBL_cu_mci_prog
)
summary(fit_AIBL_CU_MCI_PROG)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_AIBL_CU_MCI_PROG))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
```{r}
HABS_cu_mci_prog$visit_date <- as.Date(HABS_cu_mci_prog$visit_date, format = "%Y-%m-%d")
fit_HABS_CU_MCI_PROG <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = HABS_cu_mci_prog
)
summary(fit_HABS_CU_MCI_PROG)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_HABS_CU_MCI_PROG))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
# CU/MCI-AD Progression
```{r}
NACC_to_ad_prog$visit_date <- as.Date(NACC_to_ad_prog$visit_date, format = "%Y-%m-%d")
fit_NACC_TO_AD_PROG <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = NACC_to_ad_prog
)
summary(fit_NACC_TO_AD_PROG)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_NACC_TO_AD_PROG))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
```{r}
AIBL_to_ad_prog$visit_date <- as.Date(AIBL_to_ad_prog$visit_date, format = "%Y-%m-%d")
fit_AIBL_TO_AD_PROG <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = AIBL_to_ad_prog
)
summary(fit_AIBL_TO_AD_PROG)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_AIBL_TO_AD_PROG))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
```{r}
HABS_to_ad_prog$visit_date <- as.Date(HABS_to_ad_prog$visit_date, format = "%Y-%m-%d")
fit_HABS_TO_AD_PROG <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = HABS_to_ad_prog
)
summary(fit_HABS_TO_AD_PROG)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_HABS_TO_AD_PROG))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
# AD
```{r}
NACC_ad$visit_date <- as.Date(NACC_ad$visit_date, format = "%Y-%m-%d")
fit_NACC_AD <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = NACC_ad
)
summary(fit_NACC_AD)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_NACC_AD))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
```{r}
AIBL_ad$visit_date <- as.Date(AIBL_ad$visit_date, format = "%Y-%m-%d")
fit_AIBL_AD <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = AIBL_ad
)
summary(fit_AIBL_AD)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_AIBL_AD))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
```{r}
HABS_ad$visit_date <- as.Date(HABS_ad$visit_date, format = "%Y-%m-%d")
fit_HABS_AD <- lmer(
CDR ~ year_since_baseline +
age + sex + edu + APOE4 +
CVD + Endocrine + Psychiatric +
ACEi + ARB + BetaBlk + CCB + Diuretic + Metformin + Statin +
ACEi:year_since_baseline + ARB:year_since_baseline + BetaBlk:year_since_baseline +
CCB:year_since_baseline + Diuretic:year_since_baseline +
Metformin:year_since_baseline + Statin:year_since_baseline +
(1 + year_since_baseline | id),
data = HABS_ad
)
summary(fit_HABS_AD)
# BH-adjusted p-values
coef_summary <- coef(summary(fit_HABS_AD))
p_values <- coef_summary[, "Pr(>|t|)"]
p_adjusted <- p.adjust(p_values, method = "BH")
cat("nBH-adjusted p-values:n")
print(data.frame(Term = rownames(coef_summary), P.value = p_values, P.adjusted = p_adjusted))
```
# plots
```{r}
terms_all <-c(
"year_since_baseline","age","sex","edu","APOE4",
"CVD", "Endocrine", "Psychiatric",
"ACEi","ARB","BetaBlk","CCB",
"Diuretic","Metformin", "Statin",
"year_since_baseline:ACEi","year_since_baseline:ARB",
"year_since_baseline:BetaBlk","year_since_baseline:CCB",
"year_since_baseline:Diuretic", "year_since_baseline:Metformin",
"year_since_baseline:Statin"
)
terms_rename <- c(
"year_since_baseline"="Time",
"age"="Age (baseline)",
"sex"="Sex (ref: Male)",
"edu"="Education (z-scored)",
"APOE4"="APOE4 carrier (ref: No)",
"CVD"="CVD",
"Endocrine"="Endocrine disease",
"Psychiatric"="Psychiatric disease",
"ACEi"="ACEi",
"ARB"="ARBs",
"BetaBlk"="β-Blocker",
"CCB"="CCBs",
"Diuretic"="Diuretics",
"Metformin"="Metformin",
"Statin"="Statins",
"year_since_baseline:ACEi"="ACEi × Time",
"year_since_baseline:ARB"="ARBs × Time",
"year_since_baseline:BetaBlk"="β-Blocker × Time",
"year_since_baseline:CCB"="CCBs × Time",
"year_since_baseline:Diuretic"="Diuretics × Time",
"year_since_baseline:Metformin"="Metformin × Time",
"year_since_baseline:Statin"="Statins × Time"
)
```
```{r}
dt_cog = make_dt_from_fits(
fits = list(
`Overall` = list(fit_full_NACC, fit_full_AIBL, fit_full_HABS),
`Non-progression` = list(fit_NACC_NON_PROG, fit_AIBL_NON_PROG, fit_HABS_NON_PROG),
`CU-MCI progression` = list(fit_NACC_CU_MCI_PROG, fit_AIBL_CU_MCI_PROG, fit_HABS_CU_MCI_PROG),
`CU/MCI-ADem progression` = list(fit_NACC_TO_AD_PROG, fit_AIBL_TO_AD_PROG, fit_HABS_TO_AD_PROG),
`ADem-stable` = list(fit_NACC_AD, fit_AIBL_AD, fit_HABS_AD)
),
term= terms_all,
cohorts = c("NACC","AIBL","HABS"),
term_labels = terms_rename,
method = "REML",
digits = 3,
p_adjust = "none"
)
```
```{r}
png_path <- "../plots/forestplot/Figure 2.png"
dir.create(dirname(png_path), recursive = TRUE, showWarnings = FALSE)
# Expanded width to 15 in for a wider landscape layout (5 parallel group tracks)
img_w <- 15; img_h <- 10; img_dpi <- 500
if (requireNamespace("ragg", quietly = TRUE)){
ragg::agg_png(png_path, width = img_w, height = img_h,
units = "in", res = img_dpi, background = "white")
} else {
png(png_path, width = img_w * img_dpi, height = img_h * img_dpi,
res = img_dpi, bg = "white")
}
plot_forest_stratification_meta_expanded2(
dt = dt_cog,
terms = terms_all,
term_labels = terms_rename,
cohorts = c("NACC", "AIBL", "HABS"),
interaction_only = TRUE, # Set to TRUE if you only want to plot the drug * time rows
# Group tracking parameters for the 5 parallel blocks
prog_groups = c("Overall", "Non-progression", "CU-MCI progression", "CU/MCI-ADem progression", "ADem-stable"),
group_colors = c("#2F4F4F", "#BB9393", "#FEAD76", "#E64B35", "#756BB1"),
# Custom x-axis window boundaries for each of your 5 clinical group columns
xlim_cols = list(
`Overall` = c(-0.05, 0.05),
`Non-progression` = c(-0.025, 0.025),
`CU-MCI progression` = c(-0.1, 0.2),
`CU/MCI-AD progression` = c(-1, 1),
`AD` = c(-1, 0.5)
),
col_label_width = 0.08, # Space for left features/cohort labels
forest_col_frac = 0.48, # 48% forest line graph / 52% text window ratio inside columns
gap_feature = 0.60,
sub_row_h = 0.55,
cex_base = 0.90, # Sized perfectly for 5-column text spacing
ci_lwd = 2.4,
point_cex = 1.1,
diamond_height = 0.38,
# title = "Stratified Forest Plot – Cognitive Decline: Medication × Time Across Clinical Progression",
title = " ",
show_legend = TRUE,
show_vline = TRUE,
mar = c(6.5, 0.5, 5, 0.5) # Extended bottom cushion prevents legend clipping
)
dev.off()
message("Saved: ", png_path)
```

Forest_plots_CDR(cognitive).Rmd at commit 81fba02, no license · at the source

Overview

Authors: Yihan Wang¹, Benjamin Goudey^2,3, Colin L Masters⁴, Liang Jin¹, Yijun Pan^1,5, The Health and Aging Brain Study (HABS-HD) Study Team

ORCID iDs: Yijun Pan

Department of Neuroscience, School of Translational Medicine, Monash University, 99 Commercial Rd., Melbourne, VIC 3004, Australia
Australia BioCommons, The University of Melbourne, 21 Bedford Street, North Melbourne, Melbourne, VIC 3051, Australia
The ARC Training Centre in Cognitive Computing for Medical Technologies, The University of Melbourne, 700 Swanston Street, Carlton, Melbourne, VIC 3053, Australia
Alzheimer's Research Australia, The University of Western Australia, Nedlands, WA 6009, Australia
School of Health and Biomedical Sciences, RMIT University, 30 Janefield Drive, Bundoora, Melbourne, VIC 3083, Australia

Institutions: Monash University (Australia); The University of Melbourne (Australia); The University of Western Australia (Australia); RMIT University (Australia)

Journal: Age and ageing, volume 55, issue 9, article afag263

Dates: received 4 February 2026; accepted 20 July 2026; published online 14 September 2026; in print September 2026

Type: Research article · Language: English

License: CC BY-NC

Identifiers: DOI 10.1093/ageing/afag263 · PMID 42735389 · PMCID PMC13574286 · OpenAlex W7212916448

Open access: hybrid, a free copy (OpenAlex)

Status: code verified

Categories: human (organism), Alzheimer's / dementia (population)

Methods: Preprocessing, Statistics

Keywords: antihypertensive drugs, angiotensin-converting enzyme inhibitors, Alzheimer’s disease, dementia, metformin, older adults

MeSH: Alzheimer Disease*, Cardiovascular Agents*, Cognition*, Cognitive Dysfunction*, Aged, Aged, 80 and over, Angiotensin-Converting Enzyme Inhibitors, Australia, Disease Progression, Female, Humans, Hydroxymethylglutaryl-CoA Reductase Inhibitors, Male (* major topic)

Topic: Dementia and Cognitive Impairment Research (Psychiatry and Mental health, Medicine), according to OpenAlex

Funding: NIA NIH HHS (P30 AG062715, P30 AG066444, P30 AG066508, P30 AG066509, P30 AG066512, P30 AG066546, P30 AG072946, P30 AG086401, P30 AG066514, P30 AG072947, P30 AG072975, P30 AG072979, P30 AG062429, P30 AG066507, P30 AG066511, P30 AG066530, P30 AG072958, P30 AG072976, P30 AG072977, P30 AG072978, P30 AG066506, P30 AG066515, P30 AG066518, P30 AG066519, P30 AG072931, P30 AG072973, R01 AG070862, R01 AG079280, U19 AG078109, P20 AG068082, P30 AG066468, P30 AG086404, R01 AG058533, P30 AG062422, P30 AG062421, P30 AG062677, P30 AG066462, P30 AG072959, P30 AG072972, R01 AG054073); NIBIB NIH HHS (P41 EB015922)

Citations: not cited yet (Europe PMC); 49 references in the paper

Abstract

Background: People with dementia are often prescribed cardiometabolic medications; however, the impact of this medication exposome on cognitive trajectories remains unclear.

Methods: We analysed data from National Alzheimer’s Coordinating Center (n = 28 044), Australian Imaging, Biomarkers and Lifestyle study (n = 1708) and Health and Aging Brain Study: Health Disparities (n = 1425) to examine associations between seven classes of cardiometabolic medications and cognitive outcomes [CDR sum of boxes (CDR-SB)], using linear mixed-effects models with time-varying exposure. Subgroup analyses were conducted by sex, apolipoprotein E (APOE) ε4, baseline age group and four progression groups [i.e. non-progression, cognitively unimpaired-to-mild cognitive impairment (CU-to-MCI), CU/MCI-to-Alzheimer’s dementia (ADem) progression and ADem-stable group]. A meta-analysis was also performed.

Results: Angiotensin-converting enzyme inhibitors and metformin were associated with a slower cognitive decline. Diuretics showed comparable associations, although these effects were restricted to specific subgroups. Statins exhibited divergent, stage-specific associations, with inverse associations observed in the CU/MCI-to-ADem subgroup and positive associations in the ADem-stable subgroup. Angiotensin II receptor blockers were associated with a greater longitudinal increase in CDR-SB among participants in the non-progression subgroup. Calcium channel blockers showed limited subgroup-specific cognitive benefit but were also associated with an accelerated increase in CDR-SB over time. β-blockers demonstrated no consistent association with cognitive trajectories, with evidence of a lower CDR-SB slope observed only in a cohort-specific analysis.

Discussion: Future research should incorporate pharmacy-verified medication exposure, explicit dosing information and dynamically modelled trajectories, while routinely assessing baseline and follow-up cognitive outcomes in pharmacotherapy trials involving older adults.

Reproduced under the paper's license (CC BY-NC), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 5 matches between paragraphs and lines of code.

Sabrina-wyh/Multicohort-Medication-AD

License: none: the authors keep all their rights

State: the link answers, verified on 26 September 2026

Evidence: files inventoried

Commit: 81fba021b0f759d2fa0355d681c4850717e55aea, 28 July 2026

Languages: R (12), Jupyter (3), Python (1)

Size: 45 files, 16 scripts

Software Heritage: not archived

Found in: “Data availability”

Holds: README, 12 notebooks

Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation

Tools: tidyverse (12 files), ggplot2 (10 files), broom (4 files), pandas (4 files), lme4 (3 files), lmerTest (3 files), NumPy (3 files), metafor (2 files), psych (2 files), car (1 file), pheatmap (1 file), SciPy (1 file), statsmodels (1 file)

Availability: 1 check, the latest on 26 September 2026: the link answers

26 September 2026: the link answers

17 files

code/Figure 1 (and eFigures).Rmd, R, 558 lines
code/Figure S3(heatmap).Rmd, R, 154 lines
code/FigureS2(medication_proportion).Rmd, R, 183 lines
code/FigureS4(medication_age_trajectories_plots).Rmd, R, 118 lines
code/Forest_plots_CDR(baseline cognitive).Rmd, R, 371 lines, 1 match
code/Forest_plots_CDR(cognitive).Rmd, R, 553 lines, 2 matches
code/Forest_plots_CDR(main).Rmd, R, 837 lines
code/Tables.Rmd, R, 75 lines
code/med_utils.py, Python, 1,402 lines
code/medication-lmer-utility.R, R, 1,340 lines
code/plots.R, R, 3,442 lines
code/preprocessed_AIBL.ipynb, Jupyter, 393 lines
code/preprocessed_HABSHD.ipynb, Jupyter, 391 lines, 1 match
code/preprocessed_NACC.ipynb, Jupyter, 421 lines, 1 match
code/preprocessing.Rmd, R, 139 lines
saved/code/plots.R, R, 2,071 lines
README.md, Text, 1 line

At the source: github.com/Sabrina-wyh/Multicohort-Medication-AD

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
16 scripts, each with its path and the digest of its content;
5 matches between paragraphs of the paper and lines of the code (method lexical-v1);
neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data availability

Analyses and model development were conducted in R (version 4.3.1) and Python (version 3.11.11). Data requests should be made to NACC (radc.rush.edu), AIBL (https://aibl.org.au/) and HABS-HD (https://apps.unthsc.edu/itr/our). Analysis files for general modelling and validation (not linked to the original datasets) are available on GitHub at: https://github.com/Sabrina-wyh/Multicohort-Medication-AD. Additional information and materials are available upon request.

Reproduced under the paper's license (CC BY-NC), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 6 authors, 6 keywords, 13 MeSH terms, 2 funders, 44 references.

Cite

This paper

Wang, Y., Goudey, B., Masters, C. L., Jin, L., Pan, Y., & The Health and Aging Brain Study (HABS-HD) Study Team. (2026). Cardiometabolic medication exposures and cognitive outcomes in Alzheimer's disease. Age and ageing, 55(9), afag263. https://doi.org/10.1093/ageing/afag263

BibTeX

@article{wang2026cardiometabolic, author = {Wang, Yihan and Goudey, Benjamin and Masters, Colin L and Jin, Liang and Pan, Yijun and {The Health and Aging Brain Study (HABS-HD) Study Team}}, title = {{Cardiometabolic medication exposures and cognitive outcomes in Alzheimer's disease}}, journal = {Age and ageing}, year = {2026}, month = sep, volume = {55}, number = {9}, pages = {afag263}, publisher = {Oxford University Press}, issn = {0002-0729}, doi = {10.1093/ageing/afag263}, url = {https://doi.org/10.1093/ageing/afag263}, pmid = {42735389}, pmcid = {PMC13574286} }

RIS

TY - JOUR AU - Wang, Yihan AU - Goudey, Benjamin AU - Masters, Colin L AU - Jin, Liang AU - Pan, Yijun AU - The Health and Aging Brain Study (HABS-HD) Study Team TI - Cardiometabolic medication exposures and cognitive outcomes in Alzheimer's disease T2 - Age and ageing J2 - Age Ageing PY - 2026 DA - 2026/09/01 VL - 55 IS - 9 SP - afag263 SN - 0002-0729 PB - Oxford University Press DO - 10.1093/ageing/afag263 UR - https://doi.org/10.1093/ageing/afag263 LA - en ER -

CSL-JSON

{ "id": "10.1093/ageing/afag263", "type": "article-journal", "title": "Cardiometabolic medication exposures and cognitive outcomes in Alzheimer's disease", "container-title": "Age and ageing", "author": [ { "family": "Wang", "given": "Yihan" }, { "family": "Goudey", "given": "Benjamin" }, { "family": "Masters", "given": "Colin L" }, { "family": "Jin", "given": "Liang" }, { "family": "Pan", "given": "Yijun" }, { "literal": "The Health and Aging Brain Study (HABS-HD) Study Team" } ], "container-title-short": "Age Ageing", "volume": "55", "issue": "9", "page": "afag263", "DOI": "10.1093/ageing/afag263", "PMID": "42735389", "PMCID": "PMC13574286", "ISSN": "0002-0729", "publisher": "Oxford University Press", "URL": "https://doi.org/10.1093/ageing/afag263", "language": "en", "issued": { "date-parts": [ [ 2026, 9, 1 ] ] } }

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1038/s41467-026-73865-9 [code]: Histamine shapes the neurocomputational dynamics of human learning.
Journal: Nature communications
In common: metafor, car, broom, 7 other tools
[2] doi:10.1016/j.celrep.2026.117505 [code]: Impaired spatial coding and neuronal hyperactivity in the medial entorhinal cortex of aged APP knock-in mice.
Journal: Cell reports
In common: metafor, car, broom, 6 other tools, Alzheimer's / dementia
[3] doi:10.1016/j.nicl.2026.104012 [code]: Structural-functional multilayer brain network properties and outcome of combined repetitive transcranial magnetic stimulation and psychotherapy for obsessive-compulsive disorder.
Journal: NeuroImage. Clinical
In common: psych, car, lmerTest, 7 other tools
[4] doi:10.34133/csbj.0042 [code]: Using Steady-State Visual Evoked Potentials to Characterize Wide-Ranging Retinopathy Linked to <i>CRB1</i>: Implications for Clinical Trials.
Journal: Computational and structural biotechnology journal
In common: psych, car, lmerTest, 7 other tools
[5] doi:10.1162/imag.a.105 [code]: Right posterior theta reflects human parahippocampal phase resetting by salient cues during goal-directed navigation
Journal: n/a
In common: psych, car, lmerTest, 7 other tools
[6] doi:10.64898/2026.03.09.710596 [code]: Infant gut microbiomes contribute to metabolic states that impact brain function
Journal: bioRxiv (preprint)
In common: car, broom, lmerTest, 6 other tools
[7] doi:10.1038/s42003-026-10282-0 [code]: Genetic risk of Alzheimer's disease is associated with loss of brain network segregation in midlife.
Journal: Communications biology
In common: psych, car, broom, 4 other tools, Alzheimer's / dementia
[8] doi:10.1093/braincomms/fcag121 [code]: Anterior insular co-activation patterns associated with stress markers in chronic primary pain.
Journal: Brain communications
In common: psych, car, lmerTest, 6 other tools
[9] doi:10.1073/pnas.2603114123 [code]: The human hippocampus can pattern separate memories by meaning.
Journal: Proceedings of the National Academy of Sciences of the United States of America
In common: psych, broom, lmerTest, 6 other tools
[10] doi:10.1093/cercor/bhag113 [code]: Long-term reliability and stability of parameterized resting state EEG: evidence from a five-year follow-up.
Journal: Cerebral cortex (New York, N.Y. : 1991)
In common: psych, broom, lmerTest, 6 other tools

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.