OSCR

Association of alcohol use with helmet use in cyclists: analysis of the National Trauma Data Bank.

Code ↔ Paper

7 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 7 matches
  1. [1] § Methods › Study outcomes ↔ 03_Statistical_Analysis.Rmd, lines 244–311 · score 0.99 · deep vein thrombosis, acute respiratory distress, acute kidney injury, cardiac arrest, alcohol withdrawal syndrome, pulmonary embolism
  2. [2] § Results ↔ 03_Statistical_Analysis.Rmd, lines 566–624 · score 0.96 · acute respiratory distress, acute kidney injury, unplanned admission, cardiac arrest, alcohol withdrawal syndrome, unplanned intubation
  3. [3] § Methods › Statistical methods ↔ 04_Feature_Selection.ipynb, lines 90–97 · score 0.85 · fold cross validation, GridSearchCV, feature selection, logistic regression, optimize, hyperparameters
  4. [4] § Methods › Data source and patient selection ↔ 03_Statistical_Analysis.Rmd, lines 325–356 · score 0.68 · BAC screening, 0.08 %, 0.02 %, ethnicity, sex, age
  5. [5] § Results ↔ 03_Statistical_Analysis.Rmd, lines 1077–1136 · score 0.61 · AIS scores, Abdominal, Chest, pelvic, neck, Extremities
  6. [6] § Methods › Data source and patient selection ↔ 04_Feature_Selection.ipynb, lines 28–35 · score 0.55 · PROTDEV_HELMET, numeric, ethnicity, sex, age, race
  7. [7] § Results ↔ 03_Statistical_Analysis.Rmd, lines 907–934 · score 0.53 · risk factors, logistic regression, tobacco, race, disorders, wearing

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 authors' code

R Markdown · 1,277 lines · 47 KB · no license · 5 matches

  1. ---
  2. title: "Statistical_Analysis.Rmd"
  3. author: "Daniel Brock"
  4. date: "8/2/2023"
  5. output: html_document
  6. ---
  8. # 0. Importing Packages and setting working directory
  10. ```{r setup, include=FALSE}
  11. # Packages
  12. library(plyr)
  13. library(tidyverse)
  14. library(finalfit)
  15. library(broom)
  16. library(ggfortify)
  17. library(ggthemes)
  18. library(patchwork)
  19. library(magrittr)
  20. library(forcats)
  21. library(rstatix)
  22. #library(rcompanion)
  23. library(survival)
  24. library(survminer)
  25. library(readxl)
  26. library(writexl)
  27. library(pheatmap)
  28. library(zoo)
  29. library(sjPlot)
  30. library(DHARMa)
  32. # Setting working directory
  33. cwd <- getwd()
  34. ```
  36. # 1. Importing merged trauma dataset from python export
  38. ```{r}
  39. # Importing datasets
  40. trauma <- read_csv(file = paste0(cwd, "/TQP_Processed/trauma_merged.csv"))
  41. comor <- read_csv(file = paste0(cwd, "/TQP_Processed/comorbidities_merged.csv"))
  42. adverse <- read_csv(file = paste0(cwd, "/TQP_Processed/adverse_events_merged.csv"))
  43. ais <- read_csv(file = paste0(cwd, "/TQP_Processed/ais_merged.csv"))
  45. # Calculating total NTDB cases (from python file)
  46. total_cases <- 997970 + 1043736 + 1097190 + 1133053 + 1209097 + 1232956 + 1300735
  47. print(total_cases)
  48. ```
  50. ## 1.1 Viewing the data and dropping important NA rows
  52. ```{r}
  53. # Replacing quote marks in ecodes
  54. trauma$ECODE_DESC <- gsub(pattern = "\"", replacement = "", x = trauma$ECODE_DESC)
  56. # Merging trauma with comorbidities
  57. trauma <- merge(trauma, comor, on = "INC_KEY")
  59. # Merging trauma with adverse events
  60. trauma <- merge(trauma, adverse, on = "INC_KEY")
  62. # Glimpsing
  63. trauma_glimpse <- ff_glimpse(trauma)
  64. trauma_glimpse_continuous <- trauma_glimpse$Continuous
  65. trauma_glimpse_categorical <- trauma_glimpse$Categorical
  66. ```
  68. ## 1.2 Refactoring data for table 1
  70. ```{r}
  71. # Refactoring Sex
  72. trauma <- trauma %>% filter(SEX != 3) %>% mutate(
  73. sex.factor = SEX %>%
  74. factor() %>%
  75. fct_recode("Male" = "1", "Female" = "2") %>%
  76. ff_label("Sex")
  77. )
  79. # Refactoring Age
  80. trauma <- trauma %>%
  81. mutate(
  82. age.factor =
  83. AGEYEARS %>%
  84. cut(breaks = c(0,10,20,30,40,50,60,70,100), include.lowest = TRUE, right = FALSE) %>%
  85. fct_recode(
  86. "0-9" = "[0,10)",
  87. "10-19" = "[10,20)",
  88. "20-29" = "[20,30)",
  89. "30-39" = "[30,40)",
  90. "40-49" = "[40,50)",
  91. "50-59" = "[50,60)",
  92. "60-69" = "[60,70)",
  93. "70+" = "[70,100]",
  94. ) %>%
  95. ff_label("Age (years)")
  96. )
  98. # Refactoring Race
  99. trauma <- trauma %>% mutate(
  100. race.factor = case_when(
  101. ASIAN == 1 ~ "Asian",
  102. PACIFICISLANDER == 1 ~ "Pacific Islander",
  103. RACEOTHER == 1 ~ "Other",
  104. AMERICANINDIAN == 1 ~ "Native American",
  105. BLACK == 1 ~ "Black",
  106. WHITE == 1 ~ "White",
  107. RACE_NA == 1 ~ "NA",
  108. RACE_UK == 1 ~ "NA"
  109. )
  110. )
  111. trauma <- trauma %>% mutate(
  112. race.factor = na_if(race.factor, "NA") %>%
  113. factor(levels = c("White", "Black", "Asian", "Native American", "Pacific Islander", "Other")) %>%
  114. ff_label("Race")
  115. )
  117. # Refactoring Ethnicity
  118. trauma <- trauma %>% mutate(
  119. ethnicity.factor = ETHNICITY %>%
  120. factor() %>%
  121. fct_recode("Hispanic" = "1", "Not Hispanic" = "2") %>%
  122. ff_label("Ethnicity")
  123. )
  125. # Refactoring Helmet Wearing
  126. trauma <- trauma %>% mutate(
  127. helmet.factor = case_when(
  128. PROTDEV_HELMET == 1 & PROTDEV_UK != 1 ~ "Helmet",
  129. PROTDEV_HELMET == 0 & PROTDEV_UK != 1 ~ "No Helmet",
  130. PROTDEV_UK == 1 ~ "NA"
  131. )
  132. )
  133. trauma <- trauma %>% mutate(
  134. helmet.factor = na_if(helmet.factor, "NA") %>%
  135. factor(levels = c("Helmet", "No Helmet")) %>%
  136. ff_label("Helmet Use")
  137. )
  139. # Refactoring hypotension
  140. trauma <- trauma %>% mutate(
  141. SBP = SBP %>% ff_label("SBP mmHg")
  142. )
  143. trauma <- trauma %>% mutate(
  144. sbp.hypo = case_when(
  145. SBP < 90 ~ "Hypotensive",
  146. SBP >= 90 ~ "Not Hypotensive"
  147. ) %>%
  148. ff_label("SBP mmHg <90")
  149. )
  151. # Refactoring TOTALGCS
  152. trauma <- trauma %>% mutate(
  153. TOTALGCS = TOTALGCS %>% ff_label("Total GCS")
  154. )
  157. # Multiple drug use
  158. trauma$total.drugs <- trauma$DRGSCR_AMPHETAMINE + trauma$DRGSCR_BARBITURATE + trauma$DRGSCR_BENZODIAZEPINES + trauma$DRGSCR_COCAINE + trauma$DRGSCR_METHAMPHETAMINE + trauma$DRGSCR_ECSTASY + trauma$DRGSCR_METHADONE + trauma$DRGSCR_OPIOID + trauma$DRGSCR_OXYCODONE + trauma$DRGSCR_PHENCYCLIDINE + trauma$DRGSCR_TRICYCLICDEPRESS + trauma$DRGSCR_CANNABINOID + trauma$DRGSCR_OTHER
  160. # Alcohol use categories
  161. trauma <- trauma %>% mutate(
  162. alcohol.use = case_when(
  163. is.na(ALCOHOLSCREENRESULT) ~ "Not Screened",
  164. ALCOHOLSCREENRESULT < 0.02 & total.drugs == 0 ~ "Sober",
  165. ALCOHOLSCREENRESULT >= 0.02 & ALCOHOLSCREENRESULT < 0.08 & total.drugs == 0 ~ "Impaired",
  166. ALCOHOLSCREENRESULT >= 0.08 & total.drugs == 0 ~ "Intoxicated",
  167. ALCOHOLSCREENRESULT >= 0.08 & total.drugs >= 1 ~ "Multi-Drug"
  168. ) %>%
  169. factor(levels = c("Not Screened", "Sober", "Impaired", "Intoxicated", "Multi-Drug")) %>%
  170. ff_label("Alcohol")
  171. )
  173. # Refactoring Drug use
  174. trauma <- trauma %>%
  175. mutate(
  176. drug.use = case_when(
  177. DRGSCR_NONE == 1 & total.drugs < 1 & ALCOHOLSCREENRESULT < 0.02 ~ "None",
  178. DRGSCR_AMPHETAMINE == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Amphetamine",
  179. DRGSCR_BARBITURATE == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Barbiturate",
  180. DRGSCR_BENZODIAZEPINES == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Benzodiazepines",
  181. DRGSCR_COCAINE == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Cocaine",
  182. DRGSCR_METHAMPHETAMINE == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Methamphetamine",
  183. DRGSCR_ECSTASY == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Ecstasy",
  184. DRGSCR_METHADONE == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Methadone",
  185. DRGSCR_OPIOID == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Opioids",
  186. DRGSCR_OXYCODONE == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Oxycodone",
  187. DRGSCR_PHENCYCLIDINE == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Phencyclidine",
  188. DRGSCR_TRICYCLICDEPRESS == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Tricyclic Antidepressants",
  189. DRGSCR_CANNABINOID == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Cannabis",
  190. DRGSCR_OTHER == 1 & total.drugs == 1 & ALCOHOLSCREENRESULT < 0.02 ~ "Other",
  191. total.drugs >= 2 & ALCOHOLSCREEN < 0.02 ~ "Multiple Drugs",
  192. total.drugs >= 1 & ALCOHOLSCREENRESULT >= 0.02 ~ "Drugs and Alcohol",
  193. TRUE ~ NA_character_ # Add a default value
  194. ) %>%
  195. factor(levels = c("None", "Cannabis", "Amphetamine", "Opioids", "Cocaine", "Benzodiazepines", "Methamphetamine", "Oxycodone", "Barbiturate", "Phencyclidine", "Tricyclic Antidepressants", "Methadone", "Ecstasy", "Drugs and Alcohol", "Other")) %>%
  196. ff_label("Drugs")
  197. )
  199. # Refactoring length of hospital stay
  200. trauma <- trauma %>% mutate(
  201. FINALDISCHARGEDAYS = FINALDISCHARGEDAYS %>% ff_label("Length of Hospital Stay (days)")
  202. )
  204. # Refactoring length of ICU stay
  205. trauma <- trauma %>% mutate(
  206. TOTALICULOS = TOTALICULOS %>% ff_label("Length of ICU Stay (days)")
  207. )
  209. # Refactoring ISS (injury severity score)
  210. trauma <- trauma %>% mutate(
  211. ISS = ISS %>% ff_label("Injury Severity Score")
  212. )
  214. trauma <- trauma %>% mutate(
  215. ISS.factor = ISS %>%
  216. cut(breaks = c(0,3,8,15,24,75), include.lowest = TRUE, right = TRUE) %>%
  217. fct_recode(
  218. "Minor (0-3)" = "[0,3]",
  219. "Moderate (4-8)" = "(3,8]",
  220. "Serious (9-15)" = "(8,15]",
  221. "Severe (16-24)" = "(15,24]",
  222. "Critical (>25)" = "(24,75]"
  223. ) %>%
  224. ff_label("ISS Categories")
  225. )
  227. # Refactoring Comorbidities
  228. trauma <- trauma %>% mutate(
  229. CC_ALCOHOLISM = CC_ALCOHOLISM %>% ff_label("Alcoholism"),
  230. CC_SMOKING = CC_SMOKING %>% ff_label("Smoker"),
  231. CC_SUBSTANCEABUSE = CC_SUBSTANCEABUSE %>% ff_label("Substance Abuse Disorder"),
  232. CC_CHF = CC_CHF %>% ff_label("Congestive Heart Failure"),
  233. CC_RENAL = CC_RENAL %>% ff_label("End Stage Renal Disease"),
  234. CC_DIABETES = CC_DIABETES %>% ff_label("Diabetes"),
  235. CC_HYPERTENSION = CC_HYPERTENSION %>% ff_label("Hypertension"),
  236. CC_COPD = CC_COPD %>% ff_label("COPD"),
  237. CC_BLEEDING = CC_BLEEDING %>% ff_label("Bleeding Disorder"),
  238. CC_ANTICOAGULANT = CC_ANTICOAGULANT %>% ff_label("Anticoagulant Therapy"),
  239. CC_CIRRHOSIS = CC_CIRRHOSIS %>% ff_label("Cirrosis"),
  240. CC_CVA = CC_CVA %>% ff_label("Cerebrovascular Accident"),
  241. CC_MENTALPERSONALITY = CC_MENTALPERSONALITY %>% ff_label("Mental/Personality Disorder")
  242. )
  244. # Refactoring Adverse Events
  245. trauma <- trauma %>% mutate(
  246. HC_ALCOHOLWITHDRAWAL = HC_ALCOHOLWITHDRAWAL %>% ff_label("Alcohol Withdrawal Syndrome"),
  247. HC_CARDARREST = HC_CARDARREST %>% ff_label("Cardiac Arrest"),
  248. HC_DVTHROMBOSIS = HC_DVTHROMBOSIS %>% ff_label("Deep Vein Thrombosis"),
  249. HC_EMBOLISM = HC_EMBOLISM %>% ff_label("Pulmonary Embolism"),
  250. HC_KIDNEY = HC_KIDNEY %>% ff_label("Acute Kidney Injury"),
  251. HC_RESPIRATORY = HC_RESPIRATORY %>% ff_label("Acute Respiratory Distress Syndrome"),
  252. HC_SEPSIS = HC_SEPSIS %>% ff_label("Sepsis"),
  253. HC_STROKECVA = HC_STROKECVA %>% ff_label("Stroke / CVA"),
  254. HC_VAPNEUMONIA = HC_VAPNEUMONIA %>% ff_label("Hosipital-acquired Pneumonia"),
  255. HC_INTUBATION = HC_INTUBATION %>% ff_label("Unplanned Intubation"),
  256. HC_UNPLANNEDICU = HC_UNPLANNEDICU %>% ff_label("Unplanned Admission to ICU"),
  257. HC_RETURNOR = HC_RETURNOR %>% ff_label("Unplanned Return to OR")
  258. )
  260. # Refactoring mortality using pulserate - this is not mortality!
  261. #trauma <- trauma %>% mutate(
  262. # mortality.factor = case_when(
  263. # PULSERATE == 0 ~ "Death",
  264. # PULSERATE != 0 ~ "Alive"
  265. # ) %>%
  266. # factor(levels = c("Death", "Alive")) %>%
  267. # ff_label("Mortality")
  268. #)
  270. # Correctly refactoring mortality using ED and Hospital discharge disposition
  271. alive_dispositions <- c(1, 2, 3, 4, 6, 7, 8, 9, 10, 11, 12, 13, 14)
  272. trauma <- trauma %>% mutate(
  273. mortality.total = case_when(
  274. EDDISCHARGEDISPOSITION == 5 | HOSPDISCHARGEDISPOSITION == 5 ~ "Deceased",
  275. EDDISCHARGEDISPOSITION %in% alive_dispositions | HOSPDISCHARGEDISPOSITION %in% alive_dispositions ~ "Alive"
  276. ) %>%
  277. factor(levels = c("Deceased", "Alive")) %>%
  278. ff_label("Total Mortality")
  279. )
  280. trauma <- trauma %>% mutate(
  281. mortality.ED = case_when(
  282. EDDISCHARGEDISPOSITION == 5 ~ "Deceased",
  283. EDDISCHARGEDISPOSITION %in% alive_dispositions ~ "Alive"
  284. ) %>%
  285. factor(levels = c("Deceased", "Alive")) %>%
  286. ff_label("ED Mortality")
  287. )
  288. trauma <- trauma %>% mutate(
  289. mortality.hospital = case_when(
  290. HOSPDISCHARGEDISPOSITION == 5 ~ "Deceased",
  291. HOSPDISCHARGEDISPOSITION %in% alive_dispositions ~ "Alive"
  292. ) %>%
  293. factor(levels = c("Deceased", "Alive")) %>%
  294. ff_label("Hospital Mortality")
  295. )
  297. # Refactoring alcohol withdrawal syndrome
  298. trauma <- trauma %>% mutate(
  299. alc.withdraw.factor = HC_ALCOHOLWITHDRAWAL %>%
  300. factor() %>%
  301. fct_recode("Positive" = "1", "None" = "0") %>%
  302. ff_label("Alcohol Withdrawal Syndrome")
  303. )
  305. # Refactoring Year
  306. trauma <- trauma %>% mutate(
  307. year.factor = Year %>%
  308. factor() %>%
  309. ff_label("Year")
  310. )
  311. ```
  313. ```{r}
  314. # Merging with annotated injury codes and place of injury codes
  315. inj_cause <- read_xlsx("tables/injury_cause_counts_with_descriptions.xlsx")
  316. inj_cause$counts <- NULL
  317. inj_place <- read_xlsx("tables/injury_place_counts_with_descriptions.xlsx")
  318. inj_place$counts <- NULL
  320. trauma <- merge(trauma, y = inj_cause, on = "PRIMARYECODEICD10")
  321. #trauma <- merge(trauma, y = inj_place, on = "PLACEOFINJURYCODE")
  322. ```
  325. ## 1.3 Filtering for NAs in important variables
  327. ```{r}
  328. # Filtering for NAs in selected columns
  329. trauma_filt <- trauma %>% filter_at(vars(sex.factor, age.factor, race.factor, ethnicity.factor, helmet.factor), all_vars(!is.na(.)))
  330. #initial filter included: sex.factor, age.factor, race.factor, ethnicity.factor, helmet.factor, SBP, sbp.hypo, TOTALGCS, FINALDISCHARGEDAYS, ISS, ISS.factor, CC_ALCOHOLISM, CC_SMOKING, CC_SUBSTANCEABUSE, CC_CHF, CC_RENAL, CC_DIABETES, CC_HYPERTENSION, CC_COPD, alc.withdraw.factor, mortality.factor
  332. # Seeing the total number of BAC screens for people with and without other drug use
  333. bac_screens <- trauma_filt %>% mutate(
  334. alcohol.use.total = case_when(
  335. is.na(ALCOHOLSCREENRESULT) ~ "Not Screened",
  336. ALCOHOLSCREENRESULT < 0.02 ~ "Sober",
  337. ALCOHOLSCREENRESULT >= 0.02 & ALCOHOLSCREENRESULT < 0.08 ~ "Impaired",
  338. ALCOHOLSCREENRESULT >= 0.08 ~ "Intoxicated"
  339. ),
  340. alcohol.use.clean = case_when(
  341. is.na(ALCOHOLSCREENRESULT) ~ "Not Screened",
  342. ALCOHOLSCREENRESULT < 0.02 & total.drugs == 0 ~ "Sober",
  343. ALCOHOLSCREENRESULT >= 0.02 & ALCOHOLSCREENRESULT < 0.08 & total.drugs == 0 ~ "Impaired",
  344. ALCOHOLSCREENRESULT >= 0.08 & total.drugs == 0 ~ "Intoxicated",
  345. ALCOHOLSCREENRESULT >= 0.08 & total.drugs >= 0 ~ "Multi-Drug"
  346. )
  347. )
  348. bac_screens_total <- sum(!is.na(bac_screens$alcohol.use.total))
  349. bac_screens_clean <- sum(!is.na(bac_screens$alcohol.use.clean))
  350. print(paste0("Number of total BAC screens in patients with known info: ", bac_screens_total))
  351. print(paste0("Number of filtered alcohol/multi-drug BAC screens in patients with known info: ", bac_screens_clean))
  353. # Filtering for only patients with BAC screens with NO other drug use (clean)
  354. keys_to_keep <- bac_screens[!is.na(bac_screens$alcohol.use.clean), ] %>% pull(INC_KEY)
  355. trauma_filt <- trauma_filt %>% filter(INC_KEY %in% keys_to_keep)
  356. ```
  358. ```{r}
  359. # Optional export to excel
  360. #write_xlsx(trauma_filt, path = paste0(cwd, "/TQP_Processed/trauma_filtered.xlsx"))
  361. ```
  365. # 2. Demographics table for alcohol use
  367. ```{r}
  368. # Table on helmet use
  369. # Explanatory & confounding variables
  370. explanatory <- c("alcohol.use", "motor_vehicle", "sex.factor", "age.factor", "race.factor", "ethnicity.factor")
  372. # Dependent variable of interest
  373. dependent <- "helmet.factor"
  375. table1 <- trauma_filt %>%
  376. summary_factorlist(dependent, explanatory,
  377. cont = "median", p_cont_para = "kruskal.test", p_cat = "chisq", p = TRUE,
  378. add_col_totals = TRUE,
  379. na_include = TRUE)
  381. # Optional export to excel
  382. #write_xlsx(x = table1, path = "tables/table1_demographics.xlsx")
  383. ```
  385. ```{r}
  386. # Table on alcohol use groups
  387. # Explanatory & confounding variables
  388. explanatory <- c("helmet.factor", "motor_vehicle", "sex.factor", "age.factor", "race.factor", "ethnicity.factor")
  390. # Dependent variable of interest
  391. dependent <- "alcohol.use"
  393. table1.1 <- trauma_filt %>%
  394. summary_factorlist(dependent, explanatory,
  395. cont = "median", p_cont_para = "kruskal.test", p_cat = "chisq", p = TRUE,
  396. add_col_totals = TRUE,
  397. na_include = TRUE)
  398. ```
  401. ```{r}
  402. # Pairwise comparison stats with row wise fisher's exact test for Alcohol use
  403. rwf1 <- table(trauma_filt$alcohol.use, trauma_filt$helmet.factor) %>% as.data.frame.matrix()
  404. rwf1 <- rwf1[ , c(2,1)]
  406. pairwiseNominalMatrix(x = table(trauma_filt$alcohol.use, trauma_filt$helmet.factor), compare = "row", fisher = TRUE, chisq = FALSE)
  408. rwf2 <- pairwise_fisher_test(rwf1, p.adjust.method = "fdr", detailed = TRUE)
  409. chisq.test(rwf1)
  410. ```
  412. ```{r}
  413. # Pairwise comparison stats with row wise fisher's exact test for motor vehicle involvement
  414. rwf1 <- table(trauma_filt$motor_vehicle, trauma_filt$helmet.factor) %>% as.data.frame.matrix()
  415. rwf1 <- rwf1[ , c(2,1)]
  417. pairwiseNominalMatrix(x = table(trauma_filt$motor_vehicle, trauma_filt$helmet.factor), compare = "column", fisher = TRUE, chisq = FALSE)
  419. rwf2 <- row_wise_fisher_test(rwf1, p.adjust.method = "fdr", detailed = TRUE)
  420. chisq.test(rwf1)
  421. ```
  423. ```{r}
  424. # Pairwise comparison stats with row wise fisher's exact test for motor vehicle involvement
  425. rwf1 <- table(trauma_filt$alcohol.use, trauma_filt$motor_vehicle) %>% as.data.frame.matrix()
  426. rwf1$unspecified <- NULL
  427. rwf1 <- rwf1[ , c(2,1)]
  429. rwf2 <- row_wise_fisher_test(rwf1, p.adjust.method = "fdr", detailed = TRUE)
  430. rwf3 <- pairwise_fisher_test(rwf1, p.adjust.method = "fdr", detailed = TRUE)
  431. chisq.test(rwf1)
  432. ```
  434. ```{r}
  435. # Pairwise wilcox test on age
  436. pairwise.wilcox.test(trauma_filt$AGEYEARS, trauma_filt$alcohol.use, p.adjust.method = "fdr")
  438. chisq.test(table(trauma_filt$age.factor, trauma_filt$helmet.factor))
  439. ```
  441. ```{r}
  442. # Pairwise comparison stats with sex
  443. rwf1 <- table(trauma_filt$sex.factor, trauma_filt$helmet.factor) %>% as.data.frame.matrix()
  444. rwf1 <- rwf1[ , c(2,1)]
  446. rwf2 <- pairwise_fisher_test(rwf1, p.adjust.method = "fdr", detailed = TRUE)
  447. chisq.test(rwf1)
  448. ```
  450. ```{r}
  451. # Pairwise comparison stats with race
  452. rwf1 <- table(trauma_filt$race.factor, trauma_filt$helmet.factor) %>% as.data.frame.matrix()
  453. rwf1 <- rwf1[ , c(2,1)]
  455. rwf2 <- pairwise_fisher_test(rwf1, p.adjust.method = "fdr", detailed = TRUE)
  456. chisq.test(rwf1)
  457. ```
  459. ```{r}
  460. # Pairwise comparison stats with ethnicity
  461. rwf1 <- table(trauma_filt$ethnicity.factor, trauma_filt$helmet.factor) %>% as.data.frame.matrix()
  462. rwf1 <- rwf1[ , c(2,1)]
  464. rwf2 <- pairwise_fisher_test(rwf1, p.adjust.method = "fdr", detailed = TRUE)
  465. chisq.test(rwf1)
  466. ```
  472. # 3. Clinical presentation and alcohol use groups
  474. ```{r}
  475. # Explanatory & confounding variables
  476. explanatory <- c("mortality.total", "mortality.ED", "mortality.hospital", "FINALDISCHARGEDAYS", "TOTALICULOS", "ISS", "ISS.factor", "TOTALGCS", "SBP", "sbp.hypo", "HC_ALCOHOLWITHDRAWAL", "HC_CARDARREST", "HC_DVTHROMBOSIS", "HC_EMBOLISM", "HC_KIDNEY", "HC_RESPIRATORY", "HC_SEPSIS", "HC_STROKECVA", "HC_VAPNEUMONIA", "HC_INTUBATION", "HC_UNPLANNEDICU", "HC_RETURNOR")
  478. # Dependent variable of interest
  479. dependent <- "alcohol.use" #helmet.factor
  481. table2 <- trauma_filt %>%
  482. summary_factorlist(dependent, explanatory,
  483. cont = "median", p_cont_para = "kruskal.test", p_cat = "chisq", p = TRUE,
  484. add_row_totals = FALSE,
  485. na_include = TRUE)
  487. # Optional export to excel
  488. #write_csv(x = table2, path = paste0(cwd, "/tables/table2_clinical_presentation.csv"))
  489. ```
  491. ```{r}
  492. # Explanatory & confounding variables
  493. explanatory <- c("mortality.total", "mortality.ED", "mortality.hospital", "FINALDISCHARGEDAYS", "TOTALICULOS", "ISS", "ISS.factor", "TOTALGCS", "SBP", "sbp.hypo", "HC_ALCOHOLWITHDRAWAL", "HC_CARDARREST", "HC_DVTHROMBOSIS", "HC_EMBOLISM", "HC_KIDNEY", "HC_RESPIRATORY", "HC_SEPSIS", "HC_STROKECVA", "HC_VAPNEUMONIA", "HC_INTUBATION", "HC_UNPLANNEDICU", "HC_RETURNOR")
  495. # Dependent variable of interest
  496. dependent <- "helmet.factor" #helmet.factor or alcohol.use
  498. table2.1 <- trauma_filt %>%
  499. summary_factorlist(dependent, explanatory,
  500. cont = "median", p_cont_para = "kruskal.test", p_cat = "chisq", p = TRUE,
  501. add_col_totals = TRUE,
  502. na_include = TRUE)
  504. # Optional export to excel
  505. #write_xlsx(x = table2.1, path = "tables/table2_clinical_presentation.xlsx")
  506. ```
  508. ```{r}
  509. # Mortality rates in helmet groups
  511. # Total mortality
  512. rwf1 <- table(trauma_filt$mortality.total, trauma_filt$helmet.factor)
  513. rwf1 <- rwf1[ , c(2,1)]
  514. fisher.test(rwf1)
  516. # ED mortality
  517. rwf1 <- table(trauma_filt$mortality.ED, trauma_filt$helmet.factor)
  518. rwf1 <- rwf1[ , c(2,1)]
  519. fisher.test(rwf1)
  521. # Hospital mortality
  522. rwf1 <- table(trauma_filt$mortality.hospital, trauma_filt$helmet.factor)
  523. rwf1 <- rwf1[ , c(2,1)]
  524. fisher.test(rwf1)
  525. ```
  527. ```{r}
  528. # Length of hospital stay
  529. wilcox.test(x = trauma_filt$FINALDISCHARGEDAYS, g = trauma_filt$helmet.factor)
  530. med_hosp_stay1 <- trauma_filt %>% dplyr::filter(helmet.factor == "Helmet")
  531. med_hosp_stay1 <- mean(med_hosp_stay1$FINALDISCHARGEDAYS, na.rm = TRUE)
  532. med_hosp_stay2 <- trauma_filt %>% dplyr::filter(helmet.factor == "No Helmet")
  533. med_hosp_stay2 <- mean(med_hosp_stay2$FINALDISCHARGEDAYS, na.rm = TRUE)
  534. print(paste0("Mean of Helmeted hospital stay: ", med_hosp_stay1, ". Mean of Non-helmeted hospital stay: ", med_hosp_stay2))
  536. # Length of ICU stay
  537. wilcox.test(x = trauma_filt$TOTALICULOS, g = trauma_filt$helmet.factor)
  538. med_hosp_stay1 <- trauma_filt %>% dplyr::filter(helmet.factor == "Helmet")
  539. med_hosp_stay1 <- mean(med_hosp_stay1$TOTALICULOS, na.rm = TRUE)
  540. med_hosp_stay2 <- trauma_filt %>% dplyr::filter(helmet.factor == "No Helmet")
  541. med_hosp_stay2 <- mean(med_hosp_stay2$TOTALICULOS, na.rm = TRUE)
  542. print(paste0("Mean of Helmeted hospital stay: ", med_hosp_stay1, ". Mean of Non-helmeted hospital stay: ", med_hosp_stay2))
  543. ```
  546. ```{r}
  547. # ISS
  548. wilcox.test(x = trauma_filt$ISS, g = trauma_filt$helmet.factor)
  549. iss_score1 <- trauma_filt %>% dplyr::filter(helmet.factor == "Helmet")
  550. iss_score1 <- mean(iss_score1$ISS, na.rm = TRUE)
  551. iss_score2 <- trauma_filt %>% dplyr::filter(helmet.factor == "No Helmet")
  552. iss_score2 <- mean(iss_score2$ISS, na.rm = TRUE)
  553. print(paste0("Mean of Helmeted ISS: ", iss_score1, ". Mean of Non-helmeted ISS: ", iss_score2))
  555. # Pairwise fisher test for ISS categories
  556. rwf1 <- table(trauma_filt$ISS.factor, trauma_filt$helmet.factor)
  557. rwf1 <- rwf1[, c(2,1)]
  558. rwf2 <- row_wise_fisher_test(xtab = rwf1, p.adjust.method = "fdr", detailed = TRUE)
  559. ```
  561. ```{r}
  562. # Total glasgow coma scale
  563. wilcox.test(TOTALGCS ~ helmet.factor, data = trauma_filt)
  564. ```
  566. ```{r}
  567. # Clinical outcomes
  568. clin_out <- function(tab, outcome, flip = FALSE) {
  569. if (flip == TRUE) {
  570. tab <- tab[, c(2,1)]
  571. }
  572. fish <- fisher.test(tab)
  573. print(outcome)
  574. df <- data.frame("clinical_outcome" = outcome, "p_value" = fish$p.value, "OR" = fish$estimate, "conf_low" = fish$conf.int[1], "conf_high" = fish$conf.int[2])
  575. rownames(df) <- NULL
  576. return(df)
  577. }
  579. # hypotensive - change to tab switch
  580. c1 <- clin_out(tab = table(trauma_filt$sbp.hypo, trauma_filt$helmet.factor), outcome = "Hypotensive", flip = TRUE)
  582. # alcohol withdrawal syndrome
  583. c2 <- clin_out(tab = table(trauma_filt$alc.withdraw.factor, trauma_filt$helmet.factor), outcome = "Alcohol Withdrawal Sydrome")
  585. # cardiac arrest
  586. c3 <- clin_out(tab = table(trauma_filt$HC_CARDARREST, trauma_filt$helmet.factor), outcome = "Cardiac Arrest")
  588. # DVT
  589. c4 <- clin_out(tab = table(trauma_filt$HC_DVTHROMBOSIS, trauma_filt$helmet.factor), outcome = "DVT")
  591. # Pulmonary embolism
  592. c5 <- clin_out(tab = table(trauma_filt$HC_EMBOLISM, trauma_filt$helmet.factor), outcome = "Pulmonary Embolism")
  594. # AKI
  595. c6 <- clin_out(tab = table(trauma_filt$HC_KIDNEY, trauma_filt$helmet.factor), outcome = "Acute Kidney Injury")
  597. # ARDS
  598. c7 <- clin_out(tab = table(trauma_filt$HC_RESPIRATORY, trauma_filt$helmet.factor), outcome = "Acute Respiratory Distress Syndrome")
  600. # Sepsis
  601. c8 <- clin_out(tab = table(trauma_filt$HC_SEPSIS, trauma_filt$helmet.factor), outcome = "Sepsis")
  603. # Stroke / CVA
  604. c9 <- clin_out(tab = table(trauma_filt$HC_STROKECVA, trauma_filt$helmet.factor), outcome = "Stroke / CVA")
  606. # Hospital-aquired pneumonia
  607. c10 <- clin_out(tab = table(trauma_filt$HC_VAPNEUMONIA, trauma_filt$helmet.factor), outcome = "Hospital-aquired pneumonia")
  609. # Unplanned intubation
  610. c11 <- clin_out(tab = table(trauma_filt$HC_INTUBATION, trauma_filt$helmet.factor), outcome = "Unplanned Intubation")
  612. # Unplanned admission to ICU
  613. c12 <- clin_out(tab = table(trauma_filt$HC_UNPLANNEDICU, trauma_filt$helmet.factor), outcome = "Unplanned admission to ICU")
  615. # Unplanned return to OR
  616. c13 <- clin_out(tab = table(trauma_filt$HC_RETURNOR, trauma_filt$helmet.factor), outcome = "Unplanned admission to OR")
  618. # Merging
  619. c <- dplyr::bind_rows(list(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13))
  620. c$OR_CI <- paste0(round(c$OR, 2), " [", round(c$conf_low, 2), "-", round(c$conf_high, 2), "]")
  622. # Exporting to excel
  623. #write_xlsx(x = c, path = "tables/adverse_outcomes_fisher.xlsx")
  624. ```
  626. ```{r}
  627. # Sytolic blood pressure (continuous)
  628. wilcox.test(x = trauma_filt$SBP, g = trauma_filt$helmet.factor)
  629. ```
  635. # 4. Comorbidities and alcohol use groups
  637. ```{r}
  638. # Explanatory & confounding variables
  639. explanatory <- c("CC_ALCOHOLISM", "CC_SMOKING", "CC_SUBSTANCEABUSE", "CC_MENTALPERSONALITY", "CC_CHF", "CC_RENAL", "CC_DIABETES", "CC_HYPERTENSION", "CC_COPD", "CC_BLEEDING", "CC_ANTICOAGULANT", "CC_CIRRHOSIS", "CC_CVA")
  641. # Dependent variable of interest
  642. dependent <- "alcohol.use" #helmet.factor or alcohol.use
  644. table3 <- trauma_filt %>%
  645. summary_factorlist(dependent, explanatory,
  646. cont = "median", p_cont_para = "kruskal.test", p_cat = "chisq", p = TRUE,
  647. add_row_totals = FALSE,
  648. na_include = TRUE)
  650. # Optional export to excel
  651. #write_xlsx(x = table3, path = "tables/table3_comorbidities.xlsx")
  652. ```
  654. ```{r}
  655. # Explanatory & confounding variables
  656. explanatory <- c("CC_ALCOHOLISM", "CC_SMOKING", "CC_SUBSTANCEABUSE", "CC_MENTALPERSONALITY", "CC_CHF", "CC_RENAL", "CC_DIABETES", "CC_HYPERTENSION", "CC_COPD", "CC_BLEEDING", "CC_ANTICOAGULANT", "CC_CIRRHOSIS", "CC_CVA")
  658. # Dependent variable of interest
  659. dependent <- "helmet.factor" #helmet.factor or alcohol.use
  661. table3.1 <- trauma_filt %>%
  662. summary_factorlist(dependent, explanatory,
  663. cont = "median", p_cont_para = "kruskal.test", p_cat = "chisq", p = TRUE,
  664. add_col_totals = TRUE,
  665. na_include = TRUE)
  667. # Optional export to excel
  668. #write_xlsx(x = table3.1, path = "tables/table3_comorbidities.xlsx")
  669. ```
  671. ```{r}
  672. # Comorbidities fishers
  674. # alcohol withdrawal syndrome
  675. d1 <- clin_out(tab = table(trauma_filt$CC_ALCOHOLISM, trauma_filt$helmet.factor), outcome = "Alcohol Use Disorder")
  677. # smoking
  678. d2 <- clin_out(tab = table(trauma_filt$CC_SMOKING, trauma_filt$helmet.factor), outcome = "Tobacco Use Disorder")
  680. # substance use
  681. d3 <- clin_out(tab = table(trauma_filt$CC_SUBSTANCEABUSE, trauma_filt$helmet.factor), outcome = "Substance Use Disorder")
  683. # mental/personality disorder
  684. d4 <- clin_out(tab = table(trauma_filt$CC_MENTALPERSONALITY, trauma_filt$helmet.factor), outcome = "Mental/Personality Disorder")
  686. # congestive heart failure
  687. d5 <- clin_out(tab = table(trauma_filt$CC_CHF, trauma_filt$helmet.factor), outcome = "Congestive heart failure")
  689. # end stage renal disease
  690. d6 <- clin_out(tab = table(trauma_filt$CC_RENAL, trauma_filt$helmet.factor), outcome = "End stage renal disease")
  692. # diabetes
  693. d7 <- clin_out(tab = table(trauma_filt$CC_DIABETES, trauma_filt$helmet.factor), outcome = "Diabetes")
  695. # hypertension
  696. d8 <- clin_out(tab = table(trauma_filt$CC_HYPERTENSION, trauma_filt$helmet.factor), outcome = "Hypertension")
  698. # COPD
  699. d9 <- clin_out(tab = table(trauma_filt$CC_COPD, trauma_filt$helmet.factor), outcome = "COPD")
  701. # bleeding disorder
  702. d10 <- clin_out(tab = table(trauma_filt$CC_BLEEDING, trauma_filt$helmet.factor), outcome = "Bleeding disorder")
  704. # anticoagulant therapy
  705. d11 <- clin_out(tab = table(trauma_filt$CC_ANTICOAGULANT, trauma_filt$helmet.factor), outcome = "Anticoagulant therapy")
  707. # Cirrhosis
  708. d12 <- clin_out(tab = table(trauma_filt$CC_CIRRHOSIS, trauma_filt$helmet.factor), outcome = "Cirrhosis")
  710. # Cerebrovascular attack
  711. d13 <- clin_out(tab = table(trauma_filt$CC_CVA, trauma_filt$helmet.factor), outcome = "Cerebrovascular attack")
  713. # Merging
  714. d <- dplyr::bind_rows(list(d1, d2, d3, d4, d5, d6, d7, d8, d9, d10, d11, d12, d13))
  715. d$OR_CI <- paste0(round(d$OR, 2), " [", round(d$conf_low, 2), "-", round(d$conf_high, 2), "]")
  717. # Exporting to excel
  718. #write_xlsx(x = d, path = "tables/comorbidities_fisher.xlsx")
  719. ```
  723. # 5. Bargraph of helmet usage
  725. ## 5.1 Helmet usage
  727. ```{r}
  728. # Data wrangling
  729. helmet_bar_df <- as.data.frame.matrix(table(trauma_filt$alcohol.use, trauma_filt$helmet.factor))
  730. helmet_bar_df <- tibble::rownames_to_column(helmet_bar_df, var = "alcohol")
  731. helmet_bar_df$alcohol <- factor(helmet_bar_df$alcohol, levels = c("Not Screened", "Sober", "Impaired", "Intoxicated", "Multi-Drug"))
  732. helmet_bar_df$percent <- (helmet_bar_df$Helmet / (helmet_bar_df$Helmet + helmet_bar_df$`No Helmet`)) * 100
  733. helmet_bar_df$percent <- round(helmet_bar_df$percent, digits = 1)
  735. # Barplot
  736. helmet_bar_plot <- helmet_bar_df %>% ggplot(aes(x = alcohol, y = percent, fill = alcohol, color = "black")) +
  737. geom_bar(stat = "identity") +
  738. geom_text(aes(label = percent), vjust = -0.5, size = 4) +
  739. scale_y_continuous(limits = c(0, 50), expand = c(0,0)) +
  740. scale_fill_manual(values = c("black", "#84888e", "#89C5D3", "#FEDBB2", "#E8ACBD")) +
  741. scale_color_manual(values = "black") +
  742. labs(x = "BAC Intoxication Status", y = "Percent Wearing a Helmet (%)", title = "Helmet Usage") +
  743. theme_classic() +
  744. theme(plot.title = element_text(hjust = 0.5, size = 16),
  745. axis.title = element_text(size = 14),
  746. axis.text.x = element_text(size = 12, angle = 45, hjust = 1),
  747. axis.text.y = element_text(size = 12),
  748. legend.position = "none")
  749. print(helmet_bar_plot)
  751. # Optional Save
  752. #ggsave(plot = helmet_bar_plot, filename = "figures/helmet_use_barplot.pdf", width = 1400, height = 1500, dpi = 300, units = "px")
  753. ```
  755. ## 5.2 Motor vehicle involvement
  757. ```{r}
  758. # Data wrangling
  759. helmet_bar_df <- as.data.frame.matrix(table(trauma_filt$alcohol.use, trauma_filt$motor_vehicle))
  760. helmet_bar_df <- tibble::rownames_to_column(helmet_bar_df, var = "alcohol")
  761. helmet_bar_df$alcohol <- factor(helmet_bar_df$alcohol, levels = c("Not Screened", "Sober", "Impaired", "Intoxicated", "Multi-Drug"))
  762. helmet_bar_df$percent <- (helmet_bar_df$yes / (helmet_bar_df$no + helmet_bar_df$unspecified + helmet_bar_df$yes)) * 100
  763. helmet_bar_df$percent <- round(helmet_bar_df$percent, digits = 1)
  765. # Barplot
  766. helmet_bar_plot <- helmet_bar_df %>% ggplot(aes(x = alcohol, y = percent, fill = alcohol, color = "black")) +
  767. geom_bar(stat = "identity") +
  768. geom_text(aes(label = percent), vjust = -0.5, size = 4) +
  769. scale_y_continuous(limits = c(0, 55), expand = c(0,0)) +
  770. scale_fill_manual(values = c("black", "#84888e", "#89C5D3", "#FEDBB2", "#E8ACBD")) +
  771. scale_color_manual(values = "black") +
  772. labs(x = "BAC Intoxication Status", y = "Percent Wearing a Helmet (%)", title = "Helmet Usage") +
  773. theme_classic() +
  774. theme(plot.title = element_text(hjust = 0.5, size = 16),
  775. axis.title = element_text(size = 14),
  776. axis.text.x = element_text(size = 12, angle = 45, hjust = 1),
  777. axis.text.y = element_text(size = 12),
  778. legend.position = "none")
  779. print(helmet_bar_plot)
  781. # Optional Save
  782. #ggsave(plot = helmet_bar_plot, filename = "figures/motor_vehicle_barplot.png", width = 1400, height = 1500, dpi = 300, units = "px")
  783. ```
  787. # 6. Dumbbell plot for comorbidities - start here
  789. ```{r}
  790. # Function to create a comorbidity dataframe for graphing in ggplot
  791. create_comor_df <- function(comorbid_values, comorbidity_title) {
  792. comor_df <- ftable(trauma_filt$alcohol.use, comorbid_values, trauma_filt$helmet.factor)
  793. comor_df <- data.frame(expand.grid(rev(attr(comor_df, "row.vars"))), unclass(comor_df))
  794. comor_df$percent <- (comor_df$X1 / (comor_df$X1 + comor_df$X2)) * 100
  795. colnames(comor_df) <- c("comorbidity", "alcohol", "helmet", "no_helmet", "percent")
  796. comor_df$comorbidity <- plyr::mapvalues(x = comor_df$comorbidity, from = c(0, 1), to = c("no", "yes"))
  797. comor_df$comorbidity <- factor(x = comor_df$comorbidity, levels = c("yes", "no"))
  798. comor_df$alcohol <- factor(x = comor_df$alcohol, levels = c("Not Screened", "Multi-Drug", "Intoxicated", "Impaired", "Sober"))
  799. comor_df$comorbidity_name <- comorbidity_title
  800. return(comor_df)
  801. }
  802. ```
  804. ```{r}
  805. # Calculating population counts for each comorbidity x alcohol intoxication status x helmet use group to find the interaction between addictive comorbidities and alcohol consumption in helmet use rates
  806. comor_alc <- create_comor_df(comorbid_values = trauma_filt$CC_ALCOHOLISM, comorbidity_title = "Alcoholism")
  807. comor_smo <- create_comor_df(comorbid_values = trauma_filt$CC_SMOKING, comorbidity_title = "Smoking")
  808. comor_sub <- create_comor_df(comorbid_values = trauma_filt$CC_SUBSTANCEABUSE, comorbidity_title = "Substance Abuse Disorder")
  809. #comor_men <- create_comor_df(comorbid_values = trauma_filt$CC_MENTALPERSONALITY, comorbidity_title = "Mental/Personality Disorder")
  810. comor_df <- dplyr::bind_rows(comor_alc, comor_smo, comor_sub) #comor_men
  811. comor_df$comorbidity_name <- plyr::mapvalues(x = comor_df$comorbidity_name, from = c("Alcoholism", "Smoking", "Substance Abuse Disorder"), to = c("Alcohol Use Disorder", "Tobacco Use Disorder", "Substance Use Disorder"))
  812. comor_df$comorbidity_name <- factor(comor_df$comorbidity_name, levels = c("Alcohol Use Disorder", "Tobacco Use Disorder", "Substance Use Disorder"))
  813. comor_df$alcohol <- factor(comor_df$alcohol, levels = rev(c("Not Screened", "Sober", "Impaired", "Intoxicated", "Multi-Drug")))
  815. # Dumbbell Plot
  816. comor_dumb_plot <- comor_df %>% ggplot(aes(x = alcohol, y = percent)) +
  817. geom_line() +
  818. geom_point(aes(color = comorbidity), size = 3) +
  819. facet_grid(cols = vars(comorbidity_name)) +
  820. scale_x_discrete(limits = rev) +
  821. scale_color_manual(values = c("#e41a1c", "#367db7")) +
  822. labs(x = "BAC Status", y = "Percent Helmet Wearing (%)", title = "Comorbidities in Alcohol Screening Groups", color = "Comorbidity") +
  823. theme_bw() +
  824. theme(plot.title = element_text(hjust = 0.5, size = 16),
  825. axis.title = element_text(size = 14),
  826. axis.text.x = element_text(size = 10, angle = 45, hjust = 1),
  827. axis.text.y = element_text(size = 10),
  828. strip.background = element_blank(),
  829. strip.text = element_text(size = 12))
  830. comor_dumb_plot
  832. # Optional Save
  833. #ggsave(plot = comor_dumb_plot, filename = "figures/comorbidities_dumbbell.pdf", width = 3000, height = 1500, dpi = 300, units = "px")
  834. ```
  836. ```{r}
  837. # Additional Plot for dumbbell
  838. rate_decline_plot <- comor_df %>% ggplot(aes(x = alcohol, y = percent, group = comorbidity_status, color = comorbidity_status)) +
  839. geom_line() +
  840. geom_point(size = 2) +
  841. scale_x_discrete(limits = rev) +
  842. scale_color_manual(values = c("#3fe1cf", "#367db7", "#0080ff", "#010080", "#fe0903", "#b90e0d", "#e1105d", "#800000")) +
  843. labs(x = "Intoxication Status", y = "Percent Helmet Wearing (%)", title = "Comorbidities in Alcohol Screening Groups", color = "Comorbidity") +
  844. theme_bw() +
  845. theme(plot.title = element_text(hjust = 0.5, size = 16),
  846. axis.title = element_text(size = 14),
  847. axis.text.x = element_text(size = 10),
  848. axis.text.y = element_text(size = 10))
  849. rate_decline_plot
  851. # Optional Save
  852. #ggsave(plot = rate_decline_plot, filename = paste0(cwd, "/figures/rate_decline_plot.pdf"), width = 2200, height = 1500, dpi = 300, units = "px")
  853. ```
  855. ## 6.1 Statistics for dumbell plot - GLM
  857. ```{r}
  858. comor_stats_df <- trauma_filt %>% dplyr::select(ALCOHOLSCREENRESULT, alcohol.use, CC_ALCOHOLISM, CC_SMOKING, CC_SUBSTANCEABUSE, CC_MENTALPERSONALITY, helmet.factor)
  859. comor_stats_df$alcohol.use <- dplyr::recode(comor_stats_df$alcohol.use, "Not Screened" = 0, "Sober" = 1, "Impaired" = 2, "Intoxicated" = 3, "Multi-Drug" = 4)
  860. comor_stats_df$helmet.factor <- dplyr::recode(comor_stats_df$helmet.factor, "Helmet" = 0, "No Helmet" = 1)
  862. # Complete model with all interacting comorbidities
  863. cc_alc_val_glm <- glm(helmet.factor ~ alcohol.use + CC_ALCOHOLISM + CC_SMOKING + CC_SUBSTANCEABUSE, family = binomial(), data = comor_stats_df)
  864. summary(cc_alc_val_glm)
  865. print(exp(coef(cc_alc_val_glm)))
  867. cc_alc_val_glm <- glm(helmet.factor ~ alcohol.use * (CC_ALCOHOLISM + CC_SMOKING + CC_SUBSTANCEABUSE), family = binomial(), data = comor_stats_df)
  868. summary(cc_alc_val_glm)
  869. print(exp(coef(cc_alc_val_glm)))
  871. # GLM alcoholism + alcohol use
  872. cc_alc_glm <- glm(helmet.factor ~ alcohol.use * CC_ALCOHOLISM, family = binomial, data = comor_stats_df)
  873. summary(cc_alc_glm)
  874. print(exp(coef(cc_alc_glm)))
  876. # GLM smoking + alcohol use
  877. cc_smo_glm <- glm(helmet.factor ~ alcohol.use * CC_SMOKING, family = binomial, data = comor_stats_df)
  878. summary(cc_smo_glm)
  879. print(exp(coef(cc_smo_glm)))
  881. # GLM substance use disorder + alcohol use
  882. cc_sud_glm <- glm(helmet.factor ~ alcohol.use * CC_SUBSTANCEABUSE, family = binomial, data = comor_stats_df)
  883. summary(cc_sud_glm)
  884. print(exp(coef(cc_sud_glm)))
  885. ```
  887. ```{r}
  888. # Modeling model correctness with DHARMa
  889. simulationOutput <- simulateResiduals(fittedModel = cc_alc_val_glm)
  890. plot(simulationOutput)
  892. #simulationOutput <- simulateResiduals(fittedModel = cc_alc_glm)
  893. #plot(simulationOutput)
  895. #simulationOutput <- simulateResiduals(fittedModel = cc_smo_glm)
  896. #plot(simulationOutput)
  898. #simulationOutput <- simulateResiduals(fittedModel = cc_sud_glm)
  899. #plot(simulationOutput)
  901. #simulationOutput <- simulateResiduals(fittedModel = cc_mpd_glm)
  902. #plot(simulationOutput)
  903. ```
  907. # 7. Logistic Regression Analysis
  909. ```{r}
  910. # Cleaning the trauma df for logistic regression
  911. trauma_filt_lr <- trauma_filt %>%
  912. select(sex.factor, AGEYEARS, age.factor, race.factor, ethnicity.factor, ALCOHOLSCREENRESULT, alcohol.use, helmet.factor, motor_vehicle,
  913. CC_ALCOHOLISM, CC_SMOKING, CC_SUBSTANCEABUSE) %>%
  914. mutate(
  915. sex.factor = sex.factor %>% factor(levels = c("Female", "Male")) %>% ff_label("Sex"),
  916. ethnicity.factor = ethnicity.factor %>% factor(levels = c("Not Hispanic", "Hispanic")) %>% ff_label("Ethnicity"),
  917. CC_ALCOHOLISM = CC_ALCOHOLISM %>% factor() %>% fct_recode("Negative" = "0", "Positive" = "1") %>% ff_label("Alcohol Use Disorder"),
  918. CC_SMOKING = CC_SMOKING %>% factor() %>% fct_recode("Negative" = "0", "Positive" = "1") %>% ff_label("Tobacco Use Disorder"),
  919. CC_SUBSTANCEABUSE = CC_SUBSTANCEABUSE %>% factor() %>% fct_recode("Negative" = "0", "Positive" = "1") %>% ff_label("Substance Use Disorder"),
  920. AGEYEARS = AGEYEARS %>% ff_label("Age (years)")
  921. )
  923. # Performing Logistic Regression and testing different variables in the model of predicting risk factors of helmet wearing
  924. dependent <- "helmet.factor"
  925. explanatory <- c("AGEYEARS", "sex.factor", "race.factor", "ethnicity.factor", "alcohol.use", "motor_vehicle",
  926. "CC_ALCOHOLISM", "CC_SMOKING", "CC_SUBSTANCEABUSE")
  927. explanatory_multi <- c("AGEYEARS", "race.factor", "ethnicity.factor", "alcohol.use",
  928. "CC_ALCOHOLISM", "CC_SMOKING", "CC_SUBSTANCEABUSE")
  929. logistic_fit <- trauma_filt_lr %>%
  930. finalfit(dependent, explanatory, explanatory_multi, keep_models = TRUE, metrics = TRUE)
  931. logistic_fit1 <- logistic_fit[[1]] %>% as.data.frame()
  932. logistic_fit2 <- logistic_fit[[2]] %>% as.character()
  933. print(logistic_fit2)
  934. ```
  936. ```{r}
  937. # Final logisitic regression to export
  938. dependent <- "helmet.factor"
  939. explanatory <- c("AGEYEARS", "sex.factor", "race.factor", "ethnicity.factor", "alcohol.use", "motor_vehicle",
  940. "CC_ALCOHOLISM", "CC_SMOKING", "CC_SUBSTANCEABUSE")
  941. logistic_fit <- trauma_filt_lr %>%
  942. finalfit(dependent, explanatory, metrics = TRUE)
  943. logistic_fit1 <- logistic_fit[[1]] %>% as.data.frame()
  944. logistic_fit2 <- logistic_fit[[2]] %>% as.character()
  945. print(logistic_fit2)
  947. # Exporting to excel
  948. #write_xlsx(x = logistic_fit1, path = "tables/table5_logistic_regression.xlsx")
  949. ```
  951. ```{r}
  952. # Odds ratio Plot
  953. dependent <- "helmet.factor"
  954. explanatory <- c("AGEYEARS", "sex.factor", "race.factor", "ethnicity.factor", "alcohol.use", "motor_vehicle",
  955. "CC_ALCOHOLISM", "CC_SMOKING", "CC_SUBSTANCEABUSE")
  957. # Exporting a Odds ratio plot
  958. #pdf(file = "figures/logistic_regression_odds_ratio_plot.pdf", height = 6, width = 9)
  959. trauma_filt_lr %>%
  960. or_plot(dependent, explanatory,
  961. remove_ref = TRUE,
  962. breaks = c(0, 1, 2, 3, 4, 5, 6, 7, 8),
  963. table_text_size = 3.5,
  964. title_text_size = 16)
  965. #dev.off()
  966. ```
  971. # 8. Supplemental Data
  973. ## 8.1 Yearly plot to get time-data for COVID
  975. ```{r}
  976. # Function for plotting graphs
  977. line_plotter <- function(df, x_axis, y_axis, x_label, y_label, plot_title) {
  978. line_plot <- df %>%
  979. ggplot(aes(x = x_axis, y = y_axis)) +
  980. geom_line(aes(group=1), size = 1) +
  981. geom_point(size = 2) +
  982. labs(x = x_label, y = y_label, title = plot_title) +
  983. theme_classic() +
  984. theme(plot.title = element_text(hjust = 0.5, size = 22),
  985. strip.background = element_blank(),
  986. strip.text = element_text(size = 14),
  987. axis.text = element_text(size = 14),
  988. axis.title = element_text(size = 16))
  989. return(line_plot)
  990. }
  991. ```
  993. ```{r}
  994. # Getting the number of bike injuries per year
  995. bike_injury_counts <- as.data.frame(table(trauma$Year))
  996. colnames(bike_injury_counts) <- c("Year", "n_count")
  998. # Yearly bike injury frequency (from python)
  999. bike_injury_counts$ntdb_totals <- c(997970, 1043736, 1097190, 1133053, 1209097, 1232956) #2017, 2018, 2019, 2020, 2021, 2022
  1001. # Bike Injury frequency over total injuries
  1002. bike_injury_counts$bike_freq <- (bike_injury_counts$n_count / bike_injury_counts$ntdb_totals) * 100
  1004. # Plotting
  1005. bike_injury_counts_plot <- line_plotter(df = bike_injury_counts,
  1006. x_axis = bike_injury_counts$Year, y_axis = bike_injury_counts$bike_freq,
  1007. x_label = "Year", y_label = "Percent Pedalcylist Injuries",
  1008. plot_title = "Pedalcylist Trauma Frequency")
  1009. bike_injury_counts_plot
  1011. # Optional Save
  1012. #ggsave(plot = bike_injury_counts_plot, filename = paste0(cwd, "/figures/yearly_pedalcyclist_trauma_frequency.pdf"), width = 1600, height = 1500, dpi = 300, units = "px")
  1013. ```
  1015. ```{r}
  1016. # Getting the intoxication rate
  1017. intox_counts <- as.data.frame.matrix(table(trauma_filt$Year, trauma_filt$alcohol.use))
  1018. intox_counts$totals <- rowSums(intox_counts)
  1019. intox_counts$intox_freq <- ((intox_counts$Intoxicated + intox_counts$`Multi-Drug`) / intox_counts$totals) * 100 #count both intoxicated & multi-drug groups?
  1020. intox_counts <- tibble::rownames_to_column(intox_counts, var = "Year")
  1022. # Plotting
  1023. intox_freq_plot <- line_plotter(df = intox_counts,
  1024. x_axis = intox_counts$Year, y_axis = intox_counts$intox_freq,
  1025. x_label = "Year", y_label = "Percent Alcohol Intoxication",
  1026. plot_title = "Intoxication Rate per Year")
  1027. intox_freq_plot
  1029. # Optional Save
  1030. #ggsave(plot = intox_freq_plot, filename = paste0(cwd, "/figures/yearly_intoxication_frequency.pdf"), width = 1600, height = 1500, dpi = 300, units = "px")
  1031. ```
  1033. ```{r}
  1034. # Getting the Helmet Use per year
  1035. helmet_counts <- as.data.frame.matrix(table(trauma_filt$Year, trauma_filt$helmet.factor))
  1036. helmet_counts$totals <- rowSums(helmet_counts)
  1037. helmet_counts$helmet_freq <- ((helmet_counts$Helmet) / helmet_counts$totals) * 100
  1038. helmet_counts <- tibble::rownames_to_column(helmet_counts, var = "Year")
  1040. # Plotting
  1041. helmet_freq_plot <- line_plotter(df = helmet_counts,
  1042. x_axis = helmet_counts$Year, y_axis = helmet_counts$helmet_freq,
  1043. x_label = "Year", y_label = "Percent Helmet Use",
  1044. plot_title = "Helmet Use per Year")
  1045. helmet_freq_plot
  1047. # Optional Save
  1048. #ggsave(plot = helmet_freq_plot, filename = paste0(cwd, "/figures/yearly_helmet_use_frequency.pdf"), width = 1600, height = 1500, dpi = 300, units = "px")
  1049. ```
  1051. ```{r}
  1052. # Statistics on yearly intoxication rates
  1053. # Dependent variable of interest
  1054. dependent <- "year.factor"
  1055. # Explanatory & confounding variables
  1056. explanatory <- c("alcohol.use", "helmet.factor")
  1057. # Table
  1058. yearly_table <- trauma_filt %>% summary_factorlist(dependent, explanatory,
  1059. p_cat = "chisq", p = TRUE)
  1060. ```
  1064. ## 8.2 Alcohol AIS analysis
  1066. ```{r}
  1067. # Filtering for filtered INC codes
  1068. ais_filt <- ais[ais$INC_KEY %in% trauma_filt$INC_KEY, ]
  1069. ais_filt <- ais_filt[ais_filt$AISSEVERITY != 9, ] #filtering unknown severity scores
  1071. # Glimpsing
  1072. ais_glimpse <- ff_glimpse(ais_filt)
  1073. ais_glimpse_cont <- ais_glimpse[[1]]
  1074. ais_glimpse_cat <- ais_glimpse[[2]]
  1075. ```
  1077. ```{r}
  1078. # Labeling AIS scores based on alcohol intoxication levels
  1079. alc_unscreened_keys <- trauma_filt %>% filter(alcohol.use == "Not Screened") %>% pull(INC_KEY)
  1080. alc_sober_keys <- trauma_filt %>% filter(alcohol.use == "Sober") %>% pull(INC_KEY)
  1081. alc_impair_keys <- trauma_filt %>% filter(alcohol.use == "Impaired") %>% pull(INC_KEY)
  1082. alc_intox_keys <- trauma_filt %>% filter(alcohol.use == "Intoxicated") %>% pull(INC_KEY)
  1083. multi_drug_keys <- trauma_filt %>% filter(alcohol.use == "Multi-Drug") %>% pull(INC_KEY)
  1085. # Labeleding AIS scores based on helmet use
  1086. helmet_yes_keys <- trauma_filt %>% filter(helmet.factor == "Helmet") %>% pull(INC_KEY)
  1087. helmet_no_keys <- trauma_filt %>% filter(helmet.factor == "No Helmet") %>% pull(INC_KEY)
  1089. ais_filt <- ais_filt %>% mutate(
  1090. alcohol.use = case_when(
  1091. INC_KEY %in% alc_unscreened_keys ~ "Not Screened",
  1092. INC_KEY %in% alc_sober_keys ~ "Sober",
  1093. INC_KEY %in% alc_impair_keys ~ "Impaired",
  1094. INC_KEY %in% alc_intox_keys ~ "Intoxicated",
  1095. INC_KEY %in% multi_drug_keys ~ "Multi-Drug"
  1096. ),
  1097. helmet.factor = case_when(
  1098. INC_KEY %in% helmet_yes_keys ~ "Helmet",
  1099. INC_KEY %in% helmet_no_keys ~ "No Helmet"
  1100. )
  1101. )
  1103. # Filtering for NAs in selected columns
  1104. ais_filt <- ais_filt %>% filter_at(vars(AISPREDOT, AISSEVERITY, AISSEVERITY, ISSREGION, alcohol.use, helmet.factor), all_vars(!is.na(.)))
  1106. # Refactoring ISS region
  1107. ais_filt <- ais_filt %>% mutate(
  1108. ISSREGION.factor = ISSREGION %>%
  1109. factor() %>%
  1110. fct_recode(
  1111. "Head & Neck" = "1",
  1112. "Chest" = "2",
  1113. "Abdominal & Pelvic" = "3",
  1114. "Extremities" = "4",
  1115. "Face" = "5",
  1116. "External" = "6"
  1117. ) %>%
  1118. ff_label("ISS Region"),
  1120. AISSEVERITY = AISSEVERITY %>%
  1121. factor() %>%
  1122. ff_label("AIS Severity Score"),
  1124. alcohol.use = alcohol.use %>%
  1125. factor(levels = c("Not Screened", "Sober", "Impaired", "Intoxicated", "Multi-Drug")) %>%
  1126. ff_label("Alcohol Use"),
  1128. helmet.factor = helmet.factor %>%
  1129. factor(levels = c("Helmet", "No Helmet")) %>%
  1130. ff_label("Helmet")
  1131. )
  1132. ais_filt <- ais_filt %>% mutate(
  1133. ISSREGION = as.numeric(ISSREGION),
  1134. AISSEVERITY = as.numeric(AISSEVERITY)
  1135. )
  1136. ```
  1138. ```{r}
  1139. # Filters for mild and severe AIS injury scores
  1140. #ais_filt_mild <- ais_filt[ais_filt$AISSEVERITY %in% c("1,", "2"), ]
  1141. ais_filt_severe <- ais_filt[ais_filt$AISSEVERITY > 2, ]
  1143. # Explanatory & confounding variables
  1144. explanatory <- c("ISSREGION.factor", "AISSEVERITY")
  1146. # Explanatory variable of interest
  1147. dependent <- "helmet.factor" #alcohol.use or helmet.factor
  1149. table4.1 <- ais_filt %>%
  1150. summary_factorlist(dependent, explanatory,
  1151. cont = "mean", p_cont_para = "aov", p_cat = "chisq", p = TRUE, column = TRUE,
  1152. add_col_totals = TRUE,
  1153. na_include = TRUE)
  1154. table4.1$severity <- "All"
  1156. table4.2 <- ais_filt_severe %>%
  1157. summary_factorlist(dependent, explanatory,
  1158. cont = "mean", p_cont_para = "aov", p_cat = "chisq", p = TRUE,
  1159. add_col_totals = TRUE,
  1160. na_include = TRUE)
  1161. table4.2$severity <- "Severe"
  1163. table4 <- bind_rows(table4.1, table4.2)
  1165. # Optional export to excel
  1166. write_xlsx(x = table4, path = "tables/table4_ais.xlsx")
  1167. ```
  1169. ```{r}
  1170. # Fishers test for ISS region - all
  1171. rwf1 <- table(ais_filt$ISSREGION.factor, ais_filt$helmet.factor)
  1172. rwf1 <- rwf1[, c(2,1)]
  1173. chisq.test(rwf1)
  1174. rwf2 <- row_wise_fisher_test(rwf1, p.adjust.method = "fdr", detailed = TRUE)
  1175. ```
  1177. ```{r}
  1178. # Fishers test for ISS region - severe
  1179. ais_filt2 <- ais_filt[ais_filt$AISSEVERITY > 2, ]
  1180. rwf1 <- table(ais_filt2$ISSREGION.factor, ais_filt2$helmet.factor)
  1181. rwf1 <- rwf1[, c(2,1)]
  1182. chisq.test(rwf1)
  1183. rwf2 <- row_wise_fisher_test(rwf1, p.adjust.method = "fdr", detailed = TRUE)
  1184. ```
  1186. ```{r}
  1187. # AIS Severity scores
  1188. wilcox.test(AISSEVERITY ~ helmet.factor, data = ais_filt)
  1190. s <- ais_filt %>% dplyr::group_by(helmet.factor) %>%
  1191. dplyr::summarise(mean_iss = mean(AISSEVERITY))
  1192. ```
  1194. ```{r}
  1195. # Assessing poly-trauma: patients with more than one injury
  1196. ais_poly_counts <- ais_filt %>% group_by(INC_KEY, alcohol.use) %>%
  1197. dplyr::summarise(poly_trauma = n())
  1199. # Explanatory variables
  1200. explanatory <- "alcohol.use"
  1202. # Explanatory variable of interest
  1203. dependent <- "poly_trauma"
  1205. # Statistics on AIS poly-trauma counts
  1206. ais_poly_table <- ais_poly_counts %>%
  1207. summary_factorlist(dependent, explanatory,
  1208. cont = "mean", p_cont_para = "aov", p_cat = "chisq", p = TRUE)
  1210. # Assessing poly-region-trauma: patients with more than one body region injury
  1211. ais_poly_region_counts <- ais_filt %>% group_by(INC_KEY, alcohol.use, ISSREGION.factor) %>%
  1212. dplyr::summarise(poly_trauma = n())
  1213. ais_poly_region_counts <- ais_poly_region_counts %>% group_by(INC_KEY, alcohol.use) %>%
  1214. dplyr::summarise(poly_trauma = n())
  1216. # Explanatory variables
  1217. explanatory <- "alcohol.use"
  1219. # Explanatory variable of interest
  1220. dependent <- "poly_trauma"
  1222. # Statistics on AIS poly-trauma counts
  1223. ais_poly_region_table <- ais_poly_region_counts %>%
  1224. summary_factorlist(dependent, explanatory,
  1225. cont = "mean", p_cont_para = "aov", p_cat = "chisq", p = TRUE)
  1228. # Merging
  1229. ais_poly_table$ais_group <- "Total AIS Injuries"
  1230. ais_poly_region_table$ais_group <- "Number of AIS Body Region Injuries"
  1231. ais_poly <- rbind(ais_poly_table, ais_poly_region_table)
  1233. # Exporting
  1234. #write_csv(x = ais_poly, file = paste0(cwd, "/tables/ais_poly-injuries.csv"))
  1235. ```

03_Statistical_Analysis.Rmd at commit f331570, no license · at the source

Overview

Authors: Daniel C Brock1, Gaibo Yan2, Chad T Wilson3, Emanuel Narcis Husu4,5
ORCID iDs: Emanuel Narcis Husu
  1. Medical Scientist Training Program, Baylor College of Medicine, Houston, Texas, USA
  2. The Department of Family Medicine, The University of Tennessee Graduate School of Medicine, Knoxville, Tennessee, USA
  3. Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, Texas, USA
  4. H. Ben Taub Department of Physical Medicine and Rehabilitation, Baylor College of Medicine, Houston, Texas, USA
  5. Department of Clinical Sciences, Chicago Medical School, Rosalind Franklin University of Medicine and Science, North Chicago, Illinois, USA
Institutions: Baylor College of Medicine (United States); University of Tennessee at Knoxville (United States); Rosalind Franklin University of Medicine and Science (United States)
Journal: Trauma surgery & acute care open, volume 11, issue 2, article e001974
Dates: received 30 June 2025; accepted 1 June 2026; published online 26 June 2026
Type: Research article · Language: English
License: CC BY-NC
Identifiers: DOI 10.1136/tsaco-2025-001974 · PMID 42375761 · PMCID PMC13311654 · OpenAlex W7166197619
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: other condition (population), traumatic brain injury (population), clinical / translational (subfield)
Methods: Statistics, Machine learning
Keywords: helmet, alcoholism, Brain Injuries, Traumatic, Head Injuries, Closed
Topic: Injury Epidemiology and Prevention (Public Health, Environmental and Occupational Health, Medicine), according to OpenAlex
Funding: Texas Children's Hospital
Citations: cited by 1 paper (Europe PMC); 40 references in the paper
Notices: A comment on this paper has been published (42422562, from Europe PMC)

Abstract

Background: Helmet use significantly reduces the risk of traumatic brain injury among cyclists; however, the influence of alcohol use on helmet-wearing adherence remains understudied in large national data sets. The goal of this study was to investigate the relationship between alcohol use and helmet use in injured cyclists.

Methods: This study is a retrospective cross-sectional analysis of 155 766 cyclists with traumatic injuries from the National Trauma Data Bank spanning 8 years (2017–2023). Cyclists were categorized by blood alcohol concentration (BAC) as follows: not screened, sober (BAC <0.02), impaired (BAC 0.02–0.08), intoxicated (BAC ≥0.08), and multidrug intoxicated (BAC ≥0.08 plus ≥1 positive drug screen). The primary outcome was helmet usage. Secondary outcomes included mortality, hospital and intensive care unit length of stay, Injury Severity Score, and adverse events. Comorbid substance use disorders were considered as covariates.

Results: Alcohol use was associated with a dose-response decrease in helmet use. Helmet adherence was 45.9% in sober cyclists (n=51 566), compared with 19.9% in impaired (n=1592), 9.8% in intoxicated (n=8327), and 6.2% in multidrug intoxicated cyclists (n=3823; p<0.001). Alcohol use was associated with higher mortality rates (OR=2.3; p<0.001) and a greater proportion of injuries in the head and neck region (OR=1.4; p<0.001). Alcohol use when cycling was significantly associated with comorbid alcohol, tobacco, and substance use disorders (p<0.001). An interaction effect was observed between alcohol use and comorbid alcohol use disorder, where individuals with a history of alcohol use disorder exhibited a lower baseline helmet use that further declined with increasing BAC.

Conclusions: Alcohol use demonstrated a significant dose-response relationship with helmet non-usage among cyclists, particularly among those with substance use disorders. This relationship represents a critical public safety concern requiring further investigation to address its public health implications.

Level of Evidence: Retrospective Epidemiologic/Prognostic study, Level III.

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 7 matches between paragraphs and lines of code.

toofastdan117/TQP_bike_alcohol_helmets

License: none: the authors keep all their rights
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: f33157079569e2ceb6bb02961eccdb24d23a76cd, 3 September 2026
Languages: Jupyter (3), R (1)
Size: 52 files, 4 scripts
Software Heritage: not archived
Found in: the text, “Statistical methods”
Holds: README, 4 notebooks
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: NumPy (3 files), pandas (3 files), broom (1 file), patchwork (1 file), pheatmap (1 file), rstatix (1 file), scikit-learn (1 file), survival (1 file), tidyverse (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
5 files
At the source: github.com/toofastdan117/TQP_bike_alcohol_helmets

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;
  • 4 scripts, each with its path and the digest of its content;
  • 7 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 statement

Data may be obtained from a third party and are not publicly available.

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 2, 28 September 2026

  • Funding: added Texas Children's Hospital

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 4 authors, 6 keywords, 35 references, 1 integrity notice.

Cite

This paper

Brock, D. C., Yan, G., Wilson, C. T., & Husu, E. N. (2026). Association of alcohol use with helmet use in cyclists: analysis of the National Trauma Data Bank. Trauma surgery & acute care open, 11(2), e001974. https://doi.org/10.1136/tsaco-2025-001974

BibTeX

@article{brock2026association,
author = {Brock, Daniel C and Yan, Gaibo and Wilson, Chad T and Husu, Emanuel Narcis},
title = {{Association of alcohol use with helmet use in cyclists: analysis of the National Trauma Data Bank}},
journal = {Trauma surgery \& acute care open},
year = {2026},
month = jun,
volume = {11},
number = {2},
pages = {e001974},
publisher = {BMJ Publishing Group},
issn = {2397-5776},
doi = {10.1136/tsaco-2025-001974},
url = {https://doi.org/10.1136/tsaco-2025-001974},
pmid = {42375761},
pmcid = {PMC13311654}
}

RIS

TY - JOUR
AU - Brock, Daniel C
AU - Yan, Gaibo
AU - Wilson, Chad T
AU - Husu, Emanuel Narcis
TI - Association of alcohol use with helmet use in cyclists: analysis of the National Trauma Data Bank
T2 - Trauma surgery & acute care open
J2 - Trauma Surg Acute Care Open
PY - 2026
DA - 2026/06/26
VL - 11
IS - 2
SP - e001974
SN - 2397-5776
PB - BMJ Publishing Group
DO - 10.1136/tsaco-2025-001974
UR - https://doi.org/10.1136/tsaco-2025-001974
LA - en
ER -

CSL-JSON

{
"id": "10.1136/tsaco-2025-001974",
"type": "article-journal",
"title": "Association of alcohol use with helmet use in cyclists: analysis of the National Trauma Data Bank",
"container-title": "Trauma surgery & acute care open",
"author": [
{
"family": "Brock",
"given": "Daniel C"
},
{
"family": "Yan",
"given": "Gaibo"
},
{
"family": "Wilson",
"given": "Chad T"
},
{
"family": "Husu",
"given": "Emanuel Narcis"
}
],
"container-title-short": "Trauma Surg Acute Care Open",
"volume": "11",
"issue": "2",
"page": "e001974",
"DOI": "10.1136/tsaco-2025-001974",
"PMID": "42375761",
"PMCID": "PMC13311654",
"ISSN": "2397-5776",
"publisher": "BMJ Publishing Group",
"URL": "https://doi.org/10.1136/tsaco-2025-001974",
"language": "en",
"issued": {
"date-parts": [
[
2026,
6,
26
]
]
}
}

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:10.1093/bioinformatics/btag592 [code]
Network-based stratification of allele-specific expression reveals patient subgroups in Huntington's disease.
Journal: Bioinformatics (Oxford, England)
In common: survival, rstatix, broom, 6 other tools, other condition
[2] doi:10.1093/neuonc/noag128 [code]
Spatially-resolved single-cell imaging of melanoma brain metastases identifies localized immune patterns predictive of immune checkpoint blockade response.
Journal: Neuro-oncology
In common: survival, rstatix, pheatmap, 4 other tools, clinical / translational, other condition
[3] doi:10.1038/s42003-026-10957-8 [code]
Brain defence by the extracellular matrix protein Cochlin.
Journal: Communications biology
In common: survival, broom, pheatmap, 5 other tools
[4] doi:10.1038/s41586-026-10612-6 [code]
Acquired genetic and cell-state changes in IDH-mutant glioma progression.
Journal: Nature
In common: survival, broom, pheatmap, 4 other tools, other condition
[5] doi:10.1016/j.xcrm.2026.102682 [code]
TET CpG sequence-context-specific DNA demethylation shapes progression of IDH-mutant gliomas.
Journal: Cell reports. Medicine
In common: survival, broom, pheatmap, 4 other tools, other condition
[6] doi:10.1016/j.isci.2026.115657 [code]
Integration of machine learning to develop a disulfidptosis model for predicting glioma prognosis, immunotherapy response, and drug.
Journal: iScience
In common: survival, broom, pheatmap, 3 other tools, clinical / translational, other condition
[7] doi:10.1016/j.cell.2026.05.026 [code]
The critical role of the endogenous immune compartment after CAR T cell therapy in recurrent GBM.
Journal: Cell
In common: survival, rstatix, patchwork, 4 other tools, other condition
[8] doi:10.1073/pnas.2609132123 [code]
A human lysosomal storage disorder toolkit for decoding proteome landscapes in cortical-like and dopaminergic-like induced neurons.
Journal: Proceedings of the National Academy of Sciences of the United States of America
In common: rstatix, broom, pheatmap, 4 other tools
[9] doi:10.3390/ijms27093997 [code]
Coordinated Multicellular Immune Programs and Drug Targets Revealed by Single-Cell Analysis in Driver-Mutated NSCLC.
Journal: International journal of molecular sciences
In common: survival, rstatix, pheatmap, 2 other tools, other condition
[10] doi:10.1002/imt2.70163 [code]
Spatial multi-omics unveils sphingolipid metabolic reprogramming within the retinal pathological niche.
Journal: iMeta
In common: rstatix, pheatmap, patchwork, 4 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.