PPMI Analysis
Danilo Lofaro
Tuesday, October 20, 2015
Loading and pre-processing Data
Get PPMI filtered dataset and Data Dictionary from NeuroMeasures folder on Dropbox
require(repmis)## Loading required package: repmisppmi_original <- repmis::source_DropboxData("PPMI_PD_filtered.csv", "mkz3fusxnrfe54b", sep = ",",header = TRUE)## Downloading data from: https://dl.dropboxusercontent.com/s/mkz3fusxnrfe54b/PPMI_PD_filtered.csv## SHA-1 hash of the downloaded data file is:
## 2b913b1c611f60c3a2bfbd0de21c9cf326d6b035dictionary_ppmi <- repmis::source_DropboxData("Data_Dictionary.csv", "umbi4gnefezu5f4", sep = ",",header = TRUE)## Downloading data from: https://dl.dropboxusercontent.com/s/umbi4gnefezu5f4/Data_Dictionary.csv## SHA-1 hash of the downloaded data file is:
## 55793be65b64e1c325c207f6bbeff676bccc7a8appmi <- ppmi_original
Dimensions, head and structure of ppmi dataset
dim(ppmi)## [1] 397 122head(ppmi)## PATNO AGE Disease Duration MDS_UPDRS Total NP1DDS NP2FREZ VLTANIM VLTVEG
## 1 3400 39 9 50 0 1 22 24
## 2 3403 69 4 51 0 0 19 9
## 3 3406 35 7 36 0 0 15 12
## 4 3407 65 7 20 0 0 25 12
## 5 3409 63 2 43 0 0 18 15
## 6 3051 71 7 41 0 0 18 16
## VLTFRUIT AGE_ASSESS_SFTANIM DVS_SFTANIM DVT_SFTANIM PTINBOTH TMGAMBLE
## 1 17 39 11 42 1 0
## 2 10 68 10 48 1 0
## 3 16 34 8 30 1 0
## 4 11 65 13 60 1 0
## 5 14 63 9 47 1 0
## 6 13 71 9 45 3 0
## CNTRLGMB TMSEX CNTRLSEX TMBUY CNTRLBUY TMEAT CNTRLEAT TMTORACT TMTMTACT
## 1 0 0 0 0 0 1 0 0 0
## 2 0 0 0 0 0 0 0 0 0
## 3 0 0 0 0 0 0 0 0 0
## 4 0 0 0 0 0 0 0 0 0
## 5 0 0 0 0 0 0 0 0 0
## 6 0 0 0 0 0 0 0 0 0
## TMTRWD TMDISMED CNTRLDSM GENDER BIOMOM BIOMOMPD BIODAD BIODADPD FULSIB
## 1 0 0 0 0 1 0 1 0 2
## 2 0 0 0 2 1 0 1 0 3
## 3 0 0 0 2 1 0 1 0 4
## 4 0 0 0 2 1 0 1 0 1
## 5 0 0 0 2 1 0 1 0 0
## 6 0 NA NA 2 1 0 1 1 0
## FULSIBPD HAFSIB HAFSIBPD MAGPAR MAGPARPD PAGPAR PAGPARPD MATAU MATAUPD
## 1 0 0 NA 2 0 2 0 1 0
## 2 0 0 NA 2 0 2 0 4 0
## 3 0 0 0 2 0 2 1 5 0
## 4 0 1 0 2 0 2 0 3 0
## 5 NA 4 0 2 0 2 0 1 0
## 6 0 0 0 2 0 2 0 17 0
## PATAU PATAUPD KIDSNUM KIDSPD DOMSIDE CAUDATE_R CAUDATE_L PUTAMEN_R
## 1 1 0 0 0 1 1.17 1.53 0.53
## 2 1 0 3 0 2 1.64 1.19 0.65
## 3 5 0 1 0 1 1.91 2.64 0.78
## 4 2 0 2 0 1 2.27 2.50 0.90
## 5 2 0 2 0 2 2.36 1.93 0.86
## 6 2 0 3 0 2 1.73 1.93 0.95
## PUTAMEN_L Mean Striatum Mean Caudate Mean Putamen SDMTOTAL SDMTVRSN
## 1 0.49 0.9300 1.350 0.510 47 1
## 2 0.43 0.9775 1.415 0.540 32 1
## 3 1.21 1.6350 2.275 0.995 49 1
## 4 1.20 1.7175 2.385 1.050 48 1
## 5 0.69 1.4600 2.145 0.775 45 1
## 6 0.74 1.3375 1.830 0.845 37 1
## AGE_ASSESS_SDM DVSD_SDM DVT_SDM STAIAD1 STAIAD2 STAIAD3 STAIAD4 STAIAD5
## 1 39 -0.667 43.33 3 2 1 1 3
## 2 68 -1.000 40.00 2 2 2 2 3
## 3 34 -1.000 40.00 4 4 1 2 2
## 4 65 0.400 54.00 4 3 1 1 4
## 5 63 -0.300 47.00 4 4 2 2 3
## 6 71 -0.583 44.17 4 4 1 1 4
## STAIAD6 STAIAD7 STAIAD8 STAIAD9 STAIAD10 STAIAD11 STAIAD12 STAIAD13
## 1 1 2 2 1 3 3 2 1
## 2 1 1 3 3 3 3 3 2
## 3 1 1 3 1 4 3 1 1
## 4 1 2 3 1 3 4 1 1
## 5 1 1 4 1 3 4 1 1
## 6 1 1 4 1 4 4 1 1
## STAIAD14 STAIAD15 STAIAD16 STAIAD17 STAIAD18 STAIAD19 STAIAD20 STAIAD21
## 1 1 3 3 2 1 3 3 2
## 2 3 2 2 2 2 2 2 3
## 3 2 4 3 1 1 4 4 3
## 4 1 4 2 2 1 4 4 4
## 5 1 3 4 1 1 2 4 3
## 6 2 4 4 1 1 4 4 4
## STAIAD22 STAIAD23 STAIAD24 STAIAD25 STAIAD26 STAIAD27 STAIAD28 STAIAD29
## 1 3 2 1 1 2 1 2 3
## 2 2 3 2 1 2 3 2 1
## 3 3 2 3 2 2 2 3 4
## 4 1 4 1 1 3 4 2 1
## 5 2 4 1 1 2 3 2 2
## 6 1 4 1 1 3 4 1 2
## STAIAD30 STAIAD31 STAIAD32 STAIAD33 STAIAD34 STAIAD35 STAIAD36 STAIAD37
## 1 1 2 2 2 2 2 3 4
## 2 3 1 2 2 2 2 3 1
## 3 3 2 3 3 3 3 2 4
## 4 3 1 1 3 4 1 3 1
## 5 3 1 1 4 4 2 4 2
## 6 4 1 1 4 3 1 4 1
## STAIAD38 STAIAD39 STAIAD40 EDUCYRS HANDED PTCGBOTH DRMVIVID DRMAGRAC
## 1 2 3 4 18 1 1 1 1
## 2 1 3 2 18 1 1 1 1
## 3 3 2 2 18 1 1 1 1
## 4 1 4 1 16 3 3 1 0
## 5 2 4 2 14 3 1 1 0
## 6 1 4 1 18 1 3 1 0
## DRMNOCTB SLPLMBMV SLPINJUR DRMVERBL DRMFIGHT DRMUMV DRMOBJFL MVAWAKEN
## 1 0 1 0 1 1 1 1 1
## 2 0 1 0 0 0 0 0 0
## 3 0 1 0 0 0 0 1 0
## 4 0 1 0 0 0 0 0 0
## 5 0 1 0 0 1 0 0 0
## 6 0 1 1 1 1 0 0 0
## DRMREMEM SLPDSTRB STROKE HETRA PARKISM RLS NARCLPSY DEPRS EPILEPSY
## 1 1 0 0 0 1 0 0 1 0
## 2 0 1 0 0 1 0 0 0 0
## 3 1 1 0 0 1 0 0 1 0
## 4 0 1 0 0 1 0 0 0 0
## 5 0 1 0 0 1 0 0 0 0
## 6 1 1 0 0 1 0 0 0 1
## BRNINFM CNSOTH
## 1 0 0
## 2 0 0
## 3 0 0
## 4 0 0
## 5 0 0
## 6 1 0str(ppmi)## 'data.frame': 397 obs. of 122 variables:
## $ PATNO : int 3400 3403 3406 3407 3409 3051 3502 3500 3150 3451 ...
## $ AGE : num 39 69 35 65 63 71 69 49 57 54 ...
## $ Disease Duration : num 9 4 7 7 2 7 6 4 23 7 ...
## $ MDS_UPDRS Total : num 50 51 36 20 43 41 47 50 28 26 ...
## $ NP1DDS : int 0 0 0 0 0 0 0 0 0 0 ...
## $ NP2FREZ : int 1 0 0 0 0 0 0 1 0 0 ...
## $ VLTANIM : int 22 19 15 25 18 18 20 24 19 23 ...
## $ VLTVEG : int 24 9 12 12 15 16 12 11 17 11 ...
## $ VLTFRUIT : int 17 10 16 11 14 13 14 14 22 11 ...
## $ AGE_ASSESS_SFTANIM: int 39 68 34 65 63 71 69 49 57 54 ...
## $ DVS_SFTANIM : int 11 10 8 13 9 9 10 12 10 12 ...
## $ DVT_SFTANIM : int 42 48 30 60 47 45 50 54 50 51 ...
## $ PTINBOTH : int 1 1 1 1 1 3 1 1 1 1 ...
## $ TMGAMBLE : int 0 0 0 0 0 0 0 0 0 0 ...
## $ CNTRLGMB : int 0 0 0 0 0 0 0 0 0 0 ...
## $ TMSEX : int 0 0 0 0 0 0 0 0 0 0 ...
## $ CNTRLSEX : int 0 0 0 0 0 0 0 0 0 0 ...
## $ TMBUY : int 0 0 0 0 0 0 0 0 0 0 ...
## $ CNTRLBUY : int 0 0 0 0 0 0 0 0 0 0 ...
## $ TMEAT : int 1 0 0 0 0 0 0 0 0 0 ...
## $ CNTRLEAT : int 0 0 0 0 0 0 0 0 0 0 ...
## $ TMTORACT : int 0 0 0 0 0 0 0 0 0 0 ...
## $ TMTMTACT : int 0 0 0 0 0 0 0 0 1 0 ...
## $ TMTRWD : int 0 0 0 0 0 0 0 0 0 0 ...
## $ TMDISMED : int 0 0 0 0 0 NA 0 0 NA 0 ...
## $ CNTRLDSM : int 0 0 0 0 0 NA 0 0 NA 0 ...
## $ GENDER : int 0 2 2 2 2 2 2 0 1 2 ...
## $ BIOMOM : int 1 1 1 1 1 1 1 1 1 1 ...
## $ BIOMOMPD : int 0 0 0 0 0 0 0 0 0 0 ...
## $ BIODAD : int 1 1 1 1 1 1 1 1 1 1 ...
## $ BIODADPD : int 0 0 0 0 0 1 0 0 0 0 ...
## $ FULSIB : int 2 3 4 1 0 0 2 3 0 5 ...
## $ FULSIBPD : int 0 0 0 0 NA 0 0 0 NA 0 ...
## $ HAFSIB : int 0 0 0 1 4 0 1 0 0 0 ...
## $ HAFSIBPD : int NA NA 0 0 0 0 0 NA NA 0 ...
## $ MAGPAR : int 2 2 2 2 2 2 2 2 2 2 ...
## $ MAGPARPD : int 0 0 0 0 0 0 0 0 0 0 ...
## $ PAGPAR : int 2 2 2 2 2 2 2 2 2 2 ...
## $ PAGPARPD : int 0 0 1 0 0 0 0 0 0 0 ...
## $ MATAU : int 1 4 5 3 1 17 3 5 4 3 ...
## $ MATAUPD : int 0 0 0 0 0 0 0 0 0 0 ...
## $ PATAU : int 1 1 5 2 2 2 4 1 4 12 ...
## $ PATAUPD : int 0 0 0 0 0 0 0 0 0 0 ...
## $ KIDSNUM : int 0 3 1 2 2 3 2 0 0 2 ...
## $ KIDSPD : int 0 0 0 0 0 0 0 NA NA 0 ...
## $ DOMSIDE : int 1 2 1 1 2 2 1 1 2 2 ...
## $ CAUDATE_R : num 1.17 1.64 1.91 2.27 2.36 1.73 1.95 1.67 2.46 1.77 ...
## $ CAUDATE_L : num 1.53 1.19 2.64 2.5 1.93 1.93 2.51 2.05 1.94 1.57 ...
## $ PUTAMEN_R : num 0.53 0.65 0.78 0.9 0.86 0.95 0.91 0.56 0.87 1 ...
## $ PUTAMEN_L : num 0.49 0.43 1.21 1.2 0.69 0.74 0.92 0.85 0.65 0.68 ...
## $ Mean Striatum : num 0.93 0.978 1.635 1.718 1.46 ...
## $ Mean Caudate : num 1.35 1.42 2.27 2.38 2.14 ...
## $ Mean Putamen : num 0.51 0.54 0.995 1.05 0.775 0.845 0.915 0.705 0.76 0.84 ...
## $ SDMTOTAL : int 47 32 49 48 45 37 40 57 42 42 ...
## $ SDMTVRSN : int 1 1 1 1 1 1 1 1 1 1 ...
## $ AGE_ASSESS_SDM : num 39 68 34 65 63 71 69 49 57 54 ...
## $ DVSD_SDM : num -0.667 -1 -1 0.4 -0.3 ...
## $ DVT_SDM : num 43.3 40 40 54 47 ...
## $ STAIAD1 : int 3 2 4 4 4 4 4 3 3 3 ...
## $ STAIAD2 : int 2 2 4 3 4 4 4 3 4 4 ...
## $ STAIAD3 : int 1 2 1 1 2 1 2 2 1 2 ...
## $ STAIAD4 : int 1 2 2 1 2 1 1 4 1 2 ...
## $ STAIAD5 : int 3 3 2 4 3 4 4 3 3 2 ...
## $ STAIAD6 : int 1 1 1 1 1 1 1 2 1 1 ...
## $ STAIAD7 : int 2 1 1 2 1 1 1 3 1 1 ...
## $ STAIAD8 : int 2 3 3 3 4 4 4 2 3 2 ...
## $ STAIAD9 : int 1 3 1 1 1 1 1 3 1 1 ...
## $ STAIAD10 : int 3 3 4 3 3 4 4 3 3 2 ...
## $ STAIAD11 : int 3 3 3 4 4 4 4 2 3 2 ...
## $ STAIAD12 : int 2 3 1 1 1 1 2 3 1 2 ...
## $ STAIAD13 : int 1 2 1 1 1 1 1 2 1 3 ...
## $ STAIAD14 : int 1 3 2 1 1 2 1 2 2 1 ...
## $ STAIAD15 : int 3 2 4 4 3 4 4 2 3 3 ...
## $ STAIAD16 : int 3 2 3 2 4 4 4 3 2 3 ...
## $ STAIAD17 : int 2 2 1 2 1 1 2 4 1 1 ...
## $ STAIAD18 : int 1 2 1 1 1 1 1 2 1 1 ...
## $ STAIAD19 : int 3 2 4 4 2 4 4 2 4 3 ...
## $ STAIAD20 : int 3 2 4 4 4 4 4 2 4 3 ...
## $ STAIAD21 : int 2 3 3 4 3 4 3 2 4 3 ...
## $ STAIAD22 : int 3 2 3 1 2 1 1 3 1 2 ...
## $ STAIAD23 : int 2 3 2 4 4 4 4 2 3 3 ...
## $ STAIAD24 : int 1 2 3 1 1 1 3 1 1 1 ...
## $ STAIAD25 : int 1 1 2 1 1 1 1 2 1 1 ...
## $ STAIAD26 : int 2 2 2 3 2 3 2 2 4 3 ...
## $ STAIAD27 : int 1 3 2 4 3 4 4 2 3 3 ...
## $ STAIAD28 : int 2 2 3 2 2 1 1 2 1 1 ...
## $ STAIAD29 : int 3 1 4 1 2 2 2 2 1 1 ...
## $ STAIAD30 : int 1 3 3 3 3 4 4 3 4 3 ...
## $ STAIAD31 : int 2 1 2 1 1 1 1 3 2 1 ...
## $ STAIAD32 : int 2 2 3 1 1 1 1 3 2 1 ...
## $ STAIAD33 : int 2 2 3 3 4 4 4 4 4 4 ...
## $ STAIAD34 : int 2 2 3 4 4 3 4 2 2 4 ...
## $ STAIAD35 : int 2 2 3 1 2 1 1 2 1 1 ...
## $ STAIAD36 : int 3 3 2 3 4 4 4 3 4 3 ...
## $ STAIAD37 : int 4 1 4 1 2 1 2 2 2 1 ...
## $ STAIAD38 : int 2 1 3 1 2 1 1 2 2 1 ...
## $ STAIAD39 : int 3 3 2 4 4 4 4 2 4 4 ...
## $ STAIAD40 : int 4 2 2 1 2 1 1 2 1 1 ...
## $ EDUCYRS : int 18 18 18 16 14 18 16 13 13 18 ...
## [list output truncated]
Set vars (variables) names and to lowercase
names(ppmi) <- tolower(names(ppmi))
names(dictionary_ppmi) <- tolower(names(dictionary_ppmi))
Set itm_name to lowercase and find used variables in the Dictionary to check for significance
require(dplyr)
dictionary_ppmi$itm_name <- tolower(dictionary_ppmi$itm_name)
vars_ppmi <- dictionary_ppmi[dictionary_ppmi$itm_name %in% names(ppmi),c(1,2,4,5,6)]
vars_ppmi <- vars_ppmi[!(duplicated(vars_ppmi[,2])),]
names_ppmi <- data.frame(idVar=1:NCOL(ppmi),itm_name=names(ppmi))
vars_ppmi <- arrange(merge(names_ppmi,vars_ppmi,all.x = T),idVar)
vars_ppmi$itm_name <- as.character(vars_ppmi$itm_name)
Define derived variables
require(dplyr)
# Semantic Fluency
ppmi$sft <- ppmi$vltanim + ppmi$vltveg + ppmi$vltfruit
vars_ppmi <- rbind(vars_ppmi,data.frame(itm_name=names(ppmi)[NCOL(ppmi)],idVar=max(vars_ppmi$idVar)+1,mod_name="DERIVED",dscr="semantic fluency",itm_type=NA,fld_len=NA))
# QUIP
ppmi$quip <- ifelse(rowSums(ppmi[,14:15],na.rm = T)>0,1,0) + ifelse(rowSums(ppmi[,16:17],na.rm = T)>0,1,0) + ifelse(rowSums(ppmi[,18:19],na.rm = T)>0,1,0) + ifelse(rowSums(ppmi[,20:21],na.rm = T)>0,1,0) + ppmi$tmtoract + ppmi$tmtmtact + ppmi$tmtrwd
vars_ppmi <- rbind(vars_ppmi,data.frame(itm_name=names(ppmi)[NCOL(ppmi)],idVar=max(vars_ppmi$idVar)+1,mod_name="DERIVED",dscr="questionnaire for impulsive-compulsive disorders",itm_type=NA,fld_len=NA))
# Family History
ppmi$famhx <- ifelse(rowSums(ppmi[,seq(29,45,2)],na.rm = T)>0,1,0)
vars_ppmi <- rbind(vars_ppmi,data.frame(itm_name=names(ppmi)[NCOL(ppmi)],idVar=max(vars_ppmi$idVar)+1,mod_name="DERIVED",dscr="family history",itm_type=NA,fld_len=NA))
# STAI sub and total score
ppmi$staistsub <- rowSums(cbind(ppmi[,58+c(3,4,6,7,9,12,13,14,17,18)], (5-(ppmi[,58+c(1,2,5,8,10,11,15,16,19,20)]))),na.rm = T)
vars_ppmi <- rbind(vars_ppmi,data.frame(itm_name=names(ppmi)[NCOL(ppmi)],idVar=max(vars_ppmi$idVar)+1,mod_name="DERIVED",dscr="STAI - state subscore",itm_type=NA,fld_len=NA))
ppmi$staitrsub <- rowSums(cbind(ppmi[,58+c(22,24,25,28,29,31,32,35,37,38,40)], (5-(ppmi[,58+c(21,23,26,27,30,33,34,36,39)]))),na.rm = T)
vars_ppmi <- rbind(vars_ppmi,data.frame(itm_name=names(ppmi)[NCOL(ppmi)],idVar=max(vars_ppmi$idVar)+1,mod_name="DERIVED",dscr="STAI - trait subscore",itm_type=NA,fld_len=NA))
ppmi$stai <- ppmi$staistsub + ppmi$staitrsub
vars_ppmi <- rbind(vars_ppmi,data.frame(itm_name=names(ppmi)[NCOL(ppmi)],idVar=max(vars_ppmi$idVar)+1,mod_name="DERIVED",dscr="state trait anxiety total score",itm_type=NA,fld_len=NA))
# REM Sleep Disorder
ppmi$rbd <- ifelse(rowSums(ppmi[,c(114:122)],na.rm = T)>0,1,0) + rowSums(ppmi[,102:113])
vars_ppmi <- rbind(vars_ppmi,data.frame(itm_name=names(ppmi)[NCOL(ppmi)],idVar=max(vars_ppmi$idVar)+1,mod_name="DERIVED",dscr="REM sleep behavior disorder",itm_type=NA,fld_len=NA))
ppmi_bkp <- ppmi[,c(1:9,123,10:26,124,27:45,125,46:98,126:128,99:122,129)]
names(ppmi_bkp) <- gsub(" ", "_",names(ppmi_bkp))
names_ppmi_bkp <- data.frame(idVar=1:NCOL(ppmi_bkp),itm_name=names(ppmi_bkp))
vars_ppmi_bkp <- dictionary_ppmi[dictionary_ppmi$itm_name %in% names(ppmi_bkp),c(1,2,4,5,6)]
vars_ppmi_bkp <- vars_ppmi_bkp[!(duplicated(vars_ppmi_bkp[,2])),]
vars_ppmi_bkp <- arrange(merge(names_ppmi_bkp,vars_ppmi_bkp,all.x = T),idVar)
vars_ppmi_bkp$itm_name <- as.character(vars_ppmi_bkp$itm_name)
We defined freezing variable as a score > 0 on freezing and impulsive disorder as a QUIP score > 1
Build final PPMI processed dataset (Removing: Source of information variables, QUIP, Famili History and STAI single values, Creating gambling, sex, buy, eat, impOther and med variables for QUIP subsection response)
ppmi <- cbind(ppmi_bkp[,1:13], data.frame(gambling=ifelse(rowSums(ppmi_bkp[,15:16],na.rm = T)>0,1,0), sex=ifelse(rowSums(ppmi_bkp[,17:18],na.rm = T)>0,1,0), buy=ifelse(rowSums(ppmi_bkp[,19:20],na.rm = T)>0,1,0), eat=ifelse(rowSums(ppmi_bkp[,21:22],na.rm = T)>0,1,0), impOther=ifelse(rowSums(ppmi_bkp[,23:25],na.rm = T)>0,1,0), med=ifelse(rowSums(ppmi_bkp[,26:27],na.rm = T)>0,1,0)), ppmi_bkp[,c(28:29,48:61,102:106,120:NCOL(ppmi_bkp))])
ppmi$gender <- ifelse(ppmi$gender>1,2,1)
ppmi$freezing <- ifelse(ppmi$np2frez>0,1,0)
ppmi$impulsive <- ifelse(ppmi$quip>0,1,0)
names(ppmi) <- gsub(" ", "_",names(ppmi))
vars_ppmi <- dictionary_ppmi[dictionary_ppmi$itm_name %in% names(ppmi),c(1,2,4,5,6)]
vars_ppmi <- vars_ppmi[!(duplicated(vars_ppmi[,2])),]
names_ppmi <- data.frame(idVar=1:NCOL(ppmi),itm_name=names(ppmi))
vars_ppmi <- arrange(merge(names_ppmi,vars_ppmi,all.x = T),idVar)
vars_ppmi$itm_name <- as.character(vars_ppmi$itm_name)Descriptive Plots
Age, Disease duration plot
There is a positive correlation between Age, Disease Duration and MDS Score 
As expected, Age and Disease duration are significantly correalted with MDS Score at multivariate Linear Regression
summary(lm(mds_updrs_total ~ age + log(disease_duration), ppmi[ppmi$disease_duration>0,]))##
## Call:
## lm(formula = mds_updrs_total ~ age + log(disease_duration), data = ppmi[ppmi$disease_duration >
## 0, ])
##
## Residuals:
## Min 1Q Median 3Q Max
## -27.523 -9.435 -0.880 7.934 41.171
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 15.69215 4.21521 3.723 0.000226 ***
## age 0.22867 0.06659 3.434 0.000658 ***
## log(disease_duration) 1.58021 0.76732 2.059 0.040121 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 12.81 on 390 degrees of freedom
## Multiple R-squared: 0.04238, Adjusted R-squared: 0.03747
## F-statistic: 8.631 on 2 and 390 DF, p-value: 0.0002149
MDS Score vs Freezing Box Plot: patients with freezing (but not those with impulsive disorder) have significantly higher MDS score 
Semantic Fluency Score plots: No differences in Freezing or Impulsive pateints in terms os SF score as well as no correlation are found between SF score and MDS score 
The proportion of female patients affected by Freezing aas well as by Impulsive Disorder seems to be higher than males. No difference are found for Family History or Side affected at onset 
No difference in mean Striatum, Caudate, Putamen or Symbol Digit Score 
Median REM Sleep Behaviour Disorder seems higher (but not significantly) in freezing and impulsive disorder patients. 