Last Ran: 2021-06-18

Notes to do: June 15th, 2021

Add Race/Ethn.
Discuss smoking demographics discrepancies at next meeting
Test for significant differences in cognitive outcomes between those on anticholinergic medications versus not.
Model this and control for smoking, alcohol (& adjudication status)?

Task: Look for basic correlations pertaining to smoking, alcohol usage, and general demographics using health history questionnaire.

e.g., How does smoking history relate to the other measures of health history and demographics? What about alcohol intake?

Keywords: United States Pharmacopeia–National Formulary (USP–NF)

Health History Questionnaire (HHQ)

Import Data

This data is reported via the COG1 Session. We could also use the NCI Dietary questionnaire, since this includes questions about alcohol usage. Although, we'll start at Health History Questionnaire (HHQ). Tips for finding other useful variables are integrated below.

I have already used the RedCap 'Data Export Reports & Stats' tab to download this data and place it on the server under the following: "/Volumes/IGNITE_Imaging/QC_Output/R_IGNITE/RedCap/PRE/Data/HHQ"

setwd("/Volumes/IGNITE_Imaging/QC_Output/R_IGNITE/RedCap/PRE/Data/HHQ/")
FILE<-list.files("/Volumes/IGNITE_Imaging/QC_Output/R_IGNITE/RedCap/PRE/Data/HHQ/", pattern="IGNITEDatabase-IGNITEBaselineHealth*")
PATH<-"/Volumes/IGNITE_Imaging/QC_Output/R_IGNITE/RedCap/PRE/Data/HHQ/"
data<-paste(PATH, FILE, sep="")
data<-read.csv(data)
HHQraw.df<-data

Weekly Alcohol Consumption

history_beer_history - 28). How many cans of beer (12 oz.) do you have in normal week?
history_wine_history - 29). How many glasses of wine (5 oz.) do have in a normal week?
history_liquor_history - 30). How many serving of liquor (1.5 oz. shot) do have in a normal week?

Recode entries reported as...

1. Variable text entries

1.0. Typed text

Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="4/week", "4",as.character(Melted.ALC.df$value)) 
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="none", "0",as.character(Melted.ALC.df$value))

1.1. Typed fraction

exp) "1/3 <1 glass per week"

third_per_week<-1/3 
third_per_week<-as.character(third_per_week) 
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="1/3  <1 glass per week" , "0.33333" ,as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="0.5 or less", "0.5" ,as.character(Melted.ALC.df$value))

1.2. Assumed to be zero*

exp) "n/a"

Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="n/a", "0",as.character(Melted.ALC.df$value))

2. Subjective estimates outside of temporal scope

2.1 - e.g. "1/month", "2/month"

One_per_mo<-1/4 
One_per_mo<-as.character(One_per_mo) 
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="1/month", One_per_mo ,as.character(Melted.ALC.df$value))

Two_per_mo<-2/4 
Two_per_mo<-as.character(Two_per_mo) 
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="2/month", Two_per_mo ,as.character(Melted.ALC.df$value))

2.2 - e.g. "1-2/month"

One.5_per_mo<-1.5/4 
One.5_per_mo<-as.character(One.5_per_mo) 
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="1-2/month", One.5_per_mo ,as.character(Melted.ALC.df$value))

2.3 - e.g. "1/every other week"

half_per_wk<-1/2 
half_per_wk<-as.character(half_per_wk)
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="1/every other week", half_per_wk ,as.character(Melted.ALC.df$value))

3. A passive range of units with...

3.1. Narrow spread between integers

Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="0-1", "0.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="1-2", "1.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="2-3", "2.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="3-4", "3.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="4-5", "4.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="5-6", "4.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="6-7", "6.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="7-8", "6.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="8-9", "8.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="9-10", "8.5",as.character(Melted.ALC.df$value))

# Response includes string of text...
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="1-2/week", "1.5",as.character(Melted.ALC.df$value))

3.2. Moderate spread between integers

Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="0-2", "1",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="2-4", "3",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="3-5", "4",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="4-6", "5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="10-12", "11",as.character(Melted.ALC.df$value))

3.3. Large spread between integers (least reliable)*

Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="7-10", "8.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value=="2-6", "4",as.character(Melted.ALC.df$value))

4. Entries translated as dates

Entries translated as dates

exp) * "6-May"=="5-6"=> "5.5"
* "2-Jan"=="1-2"=>"1,5"
* "3-Feb"=="2-3"=>"2.5"
* "9-Aug"=="9-10"=>"9.5"
* "7-Jun" =="6-7"=>"6.5"
* "12-Oct"=="10-12"=>"11"
* "4-Feb"=="2-4"=>"3"
* "5-Mar"=="3-5"=>"4"
* "10-Jul"=="7-10"=>"8.5"

Melted.ALC.df$value<-if_else(Melted.ALC.df$value==("2-Jan") , "1.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value==("3-Feb") , "2.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value==("6-May") , "5.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value==("7-Jun") , "6.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value==("9-Aug") , "8.5",as.character(Melted.ALC.df$value))

Melted.ALC.df$value<-if_else(Melted.ALC.df$value==("4-Mar") , "3.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value==("7-Jun") , "6.5",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value==("12-Oct") , "11",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value==("4-Feb") , "3",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value==("5-Mar") , "4",as.character(Melted.ALC.df$value))
Melted.ALC.df$value<-if_else(Melted.ALC.df$value==("10-Jul") , "8.5",as.character(Melted.ALC.df$value))

5. Entries entered as '>1' => 2.

6. Entries entered as '<1'. => 1.

Alcohol Consumption Variables -

Continuous Variable 'Weekly_alc.score'

"<1" => 1
">1"=> 2

ALC.df[,3:5]<-lapply(ALC.df[,3:5], as.numeric)
ALC.df$Weekly_alc.score<-rowSums(ALC.df[,3:5])

Binary Variable 'Heavy_Drinker'

The national average is 17 per week. The Centers for Disease Control defines heavy drinking as 14 drinks per week for men and seven per week for women. A standard drink is defined as 12 ounces of beer, 5 ounces of win or 1.5 ounces of liquor. https://www.niaaa.nih.gov/alcohol-health/overview-alcohol-consumption/moderate-binge-drinking

ALC.df$Heavy_Drinker<-if_else(ALC.df$screen_gender==2 & ALC.df$Weekly_alc.score>=7, "Heavy", " ")
ALC.df$Heavy_Drinker<-if_else(ALC.df$screen_gender==1 & ALC.df$Weekly_alc.score>=14, "Heavy",as.character(ALC.df$Heavy_Drinker) )
ALC.df$Heavy_Drinker<-if_else(ALC.df$Heavy_Drinker=="Heavy", 1, 0 )

ALC.df %$%  ctable(screen_gender,Heavy_Drinker,chisq = TRUE, OR = TRUE, RR=TRUE,headings = FALSE) %>% print(method = "render")

	Heavy_Drinker
screen_gender	0				1				Total
1	124	(	87.3%	)	18	(	12.7%	)	142	(	100.0%	)
2	313	(	87.4%	)	45	(	12.6%	)	358	(	100.0%	)
Total	437	(	87.4%	)	63	(	12.6%	)	500	(	100.0%	)
Χ² = 0.0000 df = 1 p = 1.0000 O.R. (95% C.I.) = 0.99 (0.55 - 1.78) R.R. (95% C.I.) = 1.00 (0.93 - 1.08)

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

Consumption groups 'Weekly_alc.group_1'

ALC.df$Weekly_alc.group_1<-if_else(ALC.df$Weekly_alc.score==0, "none", as.character(ALC.df$Weekly_alc.score))

ALC.df$Weekly_alc.group_1<-if_else(ALC.df$Weekly_alc.score>0 &ALC.df$Weekly_alc.score<=3 , "1-3", as.character(ALC.df$Weekly_alc.group_1))

ALC.df$Weekly_alc.group_1<-if_else(ALC.df$Weekly_alc.score>3 &ALC.df$Weekly_alc.score<=5 , "3-5", as.character(ALC.df$Weekly_alc.group_1))


ALC.df$Weekly_alc.group_1<-if_else(ALC.df$Weekly_alc.score>5  &ALC.df$Weekly_alc.score<=10 , "5-10", as.character(ALC.df$Weekly_alc.group_1))


ALC.df$Weekly_alc.group_1<-if_else(ALC.df$Weekly_alc.score>10  , "10+", as.character(ALC.df$Weekly_alc.group_1))
ALC.df$Weekly_alc.group_1<-factor(ALC.df$Weekly_alc.group_1, levels=c("none", "1-3", "3-5", "5-10", "10+" ))

Consumption groups 'Weekly_alc.group_2'

0, 1-3, 3+

ALC.df$Weekly_alc.group_2<-if_else(ALC.df$Weekly_alc.score==0, "none", as.character(ALC.df$Weekly_alc.score))

ALC.df$Weekly_alc.group_2<-if_else(ALC.df$Weekly_alc.score>0 &ALC.df$Weekly_alc.score<=3 , "1-3", as.character(ALC.df$Weekly_alc.group_2))

ALC.df$Weekly_alc.group_2<-if_else(ALC.df$Weekly_alc.score>3  , "3+", as.character(ALC.df$Weekly_alc.group_2))

ALC.df$Weekly_alc.group_2<-factor(ALC.df$Weekly_alc.group_2, levels=c("none", "1-3", "3+" ))

Smoking Demographics

history_cig_history - 31). Do you currently smoke Cigarettes?
smoke_day - 31A) How many packs do you smoke per day?
yrs_smoke - 31B) For how many years have you smoked?
prior_smoke - 32) Did you previously smoke cigarettes, but quit?
packs_prior - 32A) How many packs did you previously smoke per day (on average)?
prior_smoke_yrs - 32B) For how many years did you smoke?
smoke_other - 33) Do you use any other forms of tobacco (Cigars, vaporizers, etc)?
other_type - 33A) Quantify how much of each kind you use, on average (ex. 1 cigar / month etc.)

Recode entries...

Unknown 'prior_smoke_yrs'/'smoke_other'

-9999* --> 0

Smokes.df$prior_smoke_yrs<-if_else(Smokes.df$prior_smoke_yrs=="-99993", "NA", as.character(Smokes.df$prior_smoke_yrs))
Smokes.df$prior_smoke_yrs<-as.numeric(Smokes.df$prior_smoke_yrs)
Smokes.df$smoke_other<-if_else(Smokes.df$smoke_other=="-99993", "NA", as.character(Smokes.df$smoke_other))
Smokes.df$smoke_other<-as.numeric(Smokes.df$smoke_other)

Translated as dates...

Smokes.df$packs_prior<-if_else(Smokes.df$packs_prior==("2-Jan") , "1.5",as.character(Smokes.df$packs_prior))

Typed text

Smokes.df$packs_prior<-if_else(Smokes.df$packs_prior==("1 1/2") , "1.5",as.character(Smokes.df$packs_prior))

Smokes.df$packs_prior<-if_else(Smokes.df$packs_prior==("1-1.5") , "1.25",as.character(Smokes.df$packs_prior))

Smokes.df$packs_prior<-if_else(Smokes.df$packs_prior==("1 pack per day") , "1",as.character(Smokes.df$packs_prior))

cig_day<-1/20
Smokes.df$packs_prior<-if_else(Smokes.df$packs_prior==("1 cigarette") , as.character(cig_day),as.character(Smokes.df$packs_prior))

Smokes.df$packs_prior<-if_else(Smokes.df$packs_prior==("social smoker, less than 1 pack per week") , "1",as.character(Smokes.df$packs_prior))

cig3_day<-3/20
Smokes.df$packs_prior<-if_else(Smokes.df$packs_prior==("about 3 cigarettes/day") , as.character(cig3_day),as.character(Smokes.df$packs_prior))

Entered as <1

Smokes.df$packs_prior<-if_else(Smokes.df$packs_prior==("<1") , "1",as.character(Smokes.df$packs_prior))
Smokes.df$packs_prior<-as.numeric(Smokes.df$packs_prior)


Smokes.df$packs_prior<-if_else( is.na(Smokes.df$packs_prior)==TRUE, "0",as.character(Smokes.df$packs_prior))
Smokes.df$packs_prior<-as.numeric(Smokes.df$packs_prior)

Fill incomplete smoking demographics with 0

If a person answers yes to current smoker, but missing values for smoke_day ==> 0.
Same applied for previous smokers responding NA to prior number of packs and years.

Smokes.df$prior_smoke<-as.character(Smokes.df$prior_smoke)
Smokes.df<-Smokes.df %>% 
mutate("prior_smoke"=if_else(Smokes.df$history_cig_history==1 & is.na(prior_smoke)==TRUE, "0",as.character(Smokes.df$prior_smoke)))
Smokes.df$prior_smoke<-as.numeric(Smokes.df$prior_smoke)



MISSING.DAT<-Smokes.df %>% filter(Smokes.df$history_cig_history==0 & is.na(Smokes.df$prior_smoke)==TRUE | is.na(Smokes.df$history_cig_history)==TRUE)  %>% select(record_id)

Smokes 'Other'

Smokes.df$smoke_other<-if_else(Smokes.df$other_type=="nicorette", "0", as.character(Smokes.df$smoke_other))
Smokes.df$Notes<-if_else(Smokes.df$other_type=="nicorette", "smoke_other recoded=>0", " ")

Missing Smoking Demographics Data

10499

Smoking Demographics Variables

Uncorrected 'Smoking.Status' - Never, Former, Current

Smokes.df$Smoking.Status<-if_else(Smokes.df$history_cig_history=="1" & Smokes.df$prior_smoke=="0" , "Current"," ")

Smokes.df$Smoking.Status<-if_else(Smokes.df$smoke_other=="1", "Current",as.character(Smokes.df$Smoking.Status))

Smokes.df$Smoking.Status<-if_else(Smokes.df$history_cig_history=="0" &Smokes.df$prior_smoke=="1", "Former",as.character(Smokes.df$Smoking.Status))

Smokes.df$Smoking.Status<-if_else(Smokes.df$history_cig_history=="0" & Smokes.df$prior_smoke=="0", "Never",as.character(Smokes.df$Smoking.Status))

TRUE

'Tobacco Alternative Demographics

Binary - 'Current.Smoke.factor'

Smokes.df$Current.Smoke.factor<-if_else(Smokes.df$history_cig_history=="1"|Smokes.df$smoke_other=="1" , "1","0")
Smokes.df$Current.Smoke.factor<-if_else(Smokes.df$history_cig_history=="1"|Smokes.df$smoke_other=="1" , "1","0")

former<-Smokes.df%>%filter( prior_smoke ==1)

Corrected - 'Smoking.Status'

Smokes.df$Smoking.Status<-if_else(Smokes.df$smoke_other=="1" , "Current",as.character(Smokes.df$Smoking.Status))

Smokes.df$Current.Smoke.factor<-if_else(Smokes.df$history_cig_history=="1" | Smokes.df$smoke_other=="1" , "1","0")

Former Smoker Demographics (including current alternative users)

Average Packs Prior:1.1416296
Median Packs Prior:1

Average Years Smoked:16.8925442
Median Years Smokes:15

Write out .txt file...

Alc_smoke<-Alc_smoke %>% 
  select(record_id,
         history_beer_history,history_wine_history, history_liquor_history,
         Heavy_Drinker, Weekly_alc.score, Weekly_alc.group_1, Weekly_alc.group_2 , 
         history_cig_history,smoke_day,prior_smoke, 
         prior_smoke_yrs, packs_prior, smoke_other, other_type,
         Smoking.Status,Current.Smoke.factor,
         Notes)
readr::write_csv(Alc_smoke, "/Volumes/IGNITE_Imaging/QC_Output/R_IGNITE/RedCap/PRE/Data/HHQ/OUT_DATA/Baseline_alcohol_smoke_recoded.txt")

Merge with processed/labeled RedCap database:

Baseline CIRS Data
Baseline Cognitive Variables (DSST,.., )
Baseline hru_ dat
Anything with "tscore"

OTC/RX Anticholinergics

MED.df <- readxl::read_excel("/Volumes/IGNITE_Imaging/QC_Output/R_IGNITE/RedCap/PRE/Data/Meds/IGNITE_MEDS_V3_April26.xlsx", 
    sheet = "ALL_MEDS")

MED.df$RX_ANTICHOLINERGIC<-if_else(MED.df$OTC_RX=="Prescription" & MED.df$ANTICHOLINERGIC==1,1,0)
RX_ANTICHOLINERGIC.df<-MED.df %>% filter(RX_ANTICHOLINERGIC==1)
RX_antichol_2<-RX_ANTICHOLINERGIC.df %>% filter(duplicated(RX_ANTICHOLINERGIC.df$`Record ID`)) %>% select(`Record ID`)
data$RX_ANTICHOLINERGIC<-if_else(data$record_id  %in%RX_ANTICHOLINERGIC.df$`Record ID`, 1, 0)

MED.df$OTC_ANTICHOLINERGIC<-if_else(MED.df$OTC_RX=="OTC" & MED.df$ANTICHOLINERGIC==1,1,0)
OTC_ANTICHOLINERGIC.df<-MED.df %>% filter(OTC_ANTICHOLINERGIC==1)
OTC_ANTICHOL_2<-OTC_ANTICHOLINERGIC.df %>% filter(duplicated(OTC_ANTICHOLINERGIC.df$`Record ID`)) %>% select(`Record ID`)
data$OTC_ANTICHOLINERGIC<-if_else(data$record_id  %in%OTC_ANTICHOLINERGIC.df$`Record ID`, 1, 0)

Unprocessed Baseline Medication Lists

Discrepancies between V02_data_beta, VA Classifiers, & custom fx to gather beta blocking ingredients.

VO2 Data

* 30574 V02 height still inappropriately entered.
* 20133 V02 height==NA * Thus, they're removed from sample below (until able to find the specific BMI variable used for data sharing).

data$BMI<-(data$vo2_data_weight/data$vo2_data_height/data$vo2_data_height)*10000
data$BMI<-labelled::remove_var_label(data$BMI)
data$BMI<-as.numeric(data$BMI)
class(data$BMI)="numeric" 

data$BMI_group<-if_else(data$BMI<18.5, 1, as.numeric(data$BMI)) #under
data$BMI_group<-if_else( (data$BMI_group>1 & data$BMI_group<24.9)  ,2 , as.numeric(data$BMI_group)) #healthy
data$BMI_group<-if_else(data$BMI_group<29.9  & data$BMI_group>2 ,3, as.numeric(data$BMI_group)) #over
data$BMI_group<-if_else(data$BMI_group<34.9 & data$BMI_group>3 ,4, as.numeric(data$BMI_group)) #obese I
data$BMI_group<-if_else(data$BMI_group<39.9 & data$BMI_group>4 ,5, as.numeric(data$BMI_group)) #obese II
data$BMI_group<-if_else(data$BMI_group<49.9 & data$BMI_group>5 ,6, as.numeric(data$BMI_group)) #obese III
data$BMI_group<-if_else(data$BMI_group>50 ,7, as.numeric(data$BMI_group)) #OUTLIER?
data$BMI_group<-as.character(data$BMI_group)

data$BMI_GROUP<-if_else(data$BMI_group=="1", "Underweight", data$BMI_group)
data$BMI_GROUP<-if_else(data$BMI_group=="2", "Healthy", data$BMI_GROUP)
data$BMI_GROUP<-if_else(data$BMI_group=="3", "Overweight", data$BMI_GROUP)
data$BMI_GROUP<-if_else(data$BMI_group=="4", "Obese I", data$BMI_GROUP)
data$BMI_GROUP<-if_else(data$BMI_group=="5", "Obese II", data$BMI_GROUP)
data$BMI_GROUP<-if_else(data$BMI_group=="6", "Obese III", data$BMI_GROUP)
data$BMI_GROUP<-if_else(data$BMI_group=="7", "ERROR", data$BMI_GROUP)
#data %>% filter(data$BMI_GROUP=="ERROR")
data<-data[!(data$BMI_GROUP=="ERROR"),] 
#data %>% filter(is.na(data$BMI_GROUP)==TRUE)
data$BMI_GROUP<-factor(data$BMI_GROUP, levels=c("Underweight", "Healthy", "Overweight","Obese I", "Obese II", "Obese III"))
data<-data[complete.cases(data$BMI_GROUP),] 
data$BMI_GROUP<-as.factor(data$BMI_GROUP)
data<-data[complete.cases(data$record_id),]



data$screen_race.factor<-droplevels(data$screen_race.factor)
data$race.factor<-if_else(data$screen_race.factor=="Caucasian / White",  "White",as.character(data$screen_race.factor))
data$race.factor<-if_else(data$race.factor=="African American / Black","Black", as.character(data$race.factor))
data$race.factor<-if_else(data$race.factor=="Black" | data$race.factor=="White", as.character(data$race.factor), "Other")

.

Correlograms

Dummy Coded Groups (dispersion)

V02/Physical Function/Fitness

Medications

Site Dispersion (trending/significant bolded)

Gender x Site

Race x Site

	vo2_site.factor
race.factor	Pitt				Kansas				Northeastern				Total
Black	28	(	44.4%	)	7	(	11.1%	)	28	(	44.4%	)	63	(	100.0%	)
Other	3	(	14.3%	)	8	(	38.1%	)	10	(	47.6%	)	21	(	100.0%	)
White	150	(	36.4%	)	163	(	39.6%	)	99	(	24.0%	)	412	(	100.0%	)
Total	181	(	36.5%	)	178	(	35.9%	)	137	(	27.6%	)	496	(	100.0%	)
Χ² = 27.7069 df = 4 p = .0000

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

Latina/Hispanic x Site

	vo2_site.factor
screen_race_la_his.factor	Pitt				Kansas				Northeastern				Total
Yes	2	(	11.1%	)	8	(	44.4%	)	8	(	44.4%	)	18	(	100.0%	)
No	179	(	37.4%	)	170	(	35.6%	)	129	(	27.0%	)	478	(	100.0%	)
Total	181	(	36.5%	)	178	(	35.9%	)	137	(	27.6%	)	496	(	100.0%	)
Χ² = 5.5922 df = 2 p = .0610

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

Smoking.Status x Site

	vo2_site.factor
Smoking.Status	Pitt				Kansas				Northeastern				Total
Never	85	(	32.0%	)	110	(	41.4%	)	71	(	26.7%	)	266	(	100.0%	)
Former	86	(	41.7%	)	59	(	28.6%	)	61	(	29.6%	)	206	(	100.0%	)
Current	10	(	41.7%	)	9	(	37.5%	)	5	(	20.8%	)	24	(	100.0%	)
Total	181	(	36.5%	)	178	(	35.9%	)	137	(	27.6%	)	496	(	100.0%	)
Χ² = 9.2612 df = 4 p = .0549

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

Heavy Drinkers x Site

Weekly Alcohol Groups 1 x Site

Weekly Alcohol Groups 2 x Site

Education x Site

Handedness x Site

	vo2_site.factor
hand_nih_reg.factor	Pitt				Kansas				Northeastern				Total
Right	160	(	36.3%	)	167	(	37.9%	)	114	(	25.9%	)	441	(	100.0%	)
Left	21	(	38.2%	)	11	(	20.0%	)	23	(	41.8%	)	55	(	100.0%	)
Total	181	(	36.5%	)	178	(	35.9%	)	137	(	27.6%	)	496	(	100.0%	)
Χ² = 8.9132 df = 2 p = .0116

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

Anticholinergic Status x Site

RX Anticholinergics x Site

OTC Anticholinergics x Site

Adjudication Groups x Site

	vo2_site.factor
adj_cat	Pitt				Kansas				Northeastern				Total
Performing poorly on Memory tests	21	(	31.3%	)	18	(	26.9%	)	28	(	41.8%	)	67	(	100.0%	)
Performing poorly on Non-Memory tests	23	(	29.9%	)	35	(	45.5%	)	19	(	24.7%	)	77	(	100.0%	)
Performing within normal ranges on tests	137	(	38.9%	)	125	(	35.5%	)	90	(	25.6%	)	352	(	100.0%	)
Total	181	(	36.5%	)	178	(	35.9%	)	137	(	27.6%	)	496	(	100.0%	)
Χ² = 11.1276 df = 4 p = .0252

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

term	estimate	std.error	statistic	p.value
vo2_age	0.3	0.1	3.56	0.00
vo2_sex	4.6	0.8	6.10	0.00
ses_year_education_2	0.8	0.1	5.49	0.00
vo2_site.factorPitt	-11.4	8.0	-1.43	0.15
vo2_site.factorKansas	-9.7	8.0	-1.21	0.23
vo2_site.factorNortheastern	-11.3	7.9	-1.42	0.16
vo2_site.factorPitt:adj_catPerforming poorly on Non-Memory tests	8.4	2.3	3.73	0.00
vo2_site.factorKansas:adj_catPerforming poorly on Non-Memory tests	6.5	2.2	2.98	0.00
vo2_site.factorNortheastern:adj_catPerforming poorly on Non-Memory tests	4.7	2.2	2.11	0.04
vo2_site.factorPitt:adj_catPerforming within normal ranges on tests	11.2	1.8	6.39	0.00
vo2_site.factorKansas:adj_catPerforming within normal ranges on tests	9.3	1.9	4.92	0.00
vo2_site.factorNortheastern:adj_catPerforming within normal ranges on tests	11.2	1.6	6.87	0.00

Adjudication Outcomes

Smoking Status x Adjudication Groups

OTC Anticholinergic x Adjudication Groups

RX Anticholinergic x Adjudication Groups

Smoking Status x Adjudication Groups

Education x Adjudication Groups

	ses_year_education_2.factor
adj_cat	High School 09				High School 10				High School 11				High School 12				College 13				College 14				College 15				College 16				Graduate School 17				Graduate School 18				Graduate School 19				Graduate School 20+				Total
Performing poorly on Memory tests	0	(	0.0%	)	0	(	0.0%	)	0	(	0.0%	)	9	(	13.4%	)	4	(	6.0%	)	6	(	9.0%	)	1	(	1.5%	)	15	(	22.4%	)	4	(	6.0%	)	13	(	19.4%	)	2	(	3.0%	)	13	(	19.4%	)	67	(	100.0%	)
Performing poorly on Non-Memory tests	0	(	0.0%	)	0	(	0.0%	)	1	(	1.3%	)	7	(	9.1%	)	0	(	0.0%	)	4	(	5.2%	)	1	(	1.3%	)	23	(	29.9%	)	6	(	7.8%	)	18	(	23.4%	)	5	(	6.5%	)	12	(	15.6%	)	77	(	100.0%	)
Performing within normal ranges on tests	1	(	0.3%	)	3	(	0.9%	)	1	(	0.3%	)	22	(	6.2%	)	19	(	5.4%	)	31	(	8.8%	)	13	(	3.7%	)	86	(	24.4%	)	22	(	6.2%	)	66	(	18.8%	)	26	(	7.4%	)	62	(	17.6%	)	352	(	100.0%	)
Total	1	(	0.2%	)	3	(	0.6%	)	2	(	0.4%	)	38	(	7.7%	)	23	(	4.6%	)	41	(	8.3%	)	15	(	3.0%	)	124	(	25.0%	)	32	(	6.5%	)	97	(	19.6%	)	33	(	6.7%	)	87	(	17.5%	)	496	(	100.0%	)
Χ² = 18.5581 df = 22 p = .6724

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

Smoking Status x Adjudication Groups

Weekly Alcohol Groups x Adjudication Groups

Heavy Drinkers x Adjudication Groups

	Heavy_Drinker
adj_cat	0				1				Total
Performing poorly on Memory tests	54	(	80.6%	)	13	(	19.4%	)	67	(	100.0%	)
Performing poorly on Non-Memory tests	69	(	89.6%	)	8	(	10.4%	)	77	(	100.0%	)
Performing within normal ranges on tests	310	(	88.1%	)	42	(	11.9%	)	352	(	100.0%	)
Total	433	(	87.3%	)	63	(	12.7%	)	496	(	100.0%	)
Χ² = 3.2729 df = 2 p = .1947

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

Gender x Adjudication Groups

Demographics

Gender x Education

	ses_year_education_2.factor
vo2_sex	High School 09				High School 10				High School 11				High School 12				College 13				College 14				College 15				College 16				Graduate School 17				Graduate School 18				Graduate School 19				Graduate School 20+				Total
1	0	(	0.0%	)	0	(	0.0%	)	0	(	0.0%	)	7	(	5.0%	)	4	(	2.8%	)	11	(	7.8%	)	6	(	4.3%	)	36	(	25.5%	)	10	(	7.1%	)	25	(	17.7%	)	5	(	3.5%	)	37	(	26.2%	)	141	(	100.0%	)
2	1	(	0.3%	)	3	(	0.8%	)	2	(	0.6%	)	31	(	8.7%	)	19	(	5.4%	)	30	(	8.5%	)	9	(	2.5%	)	88	(	24.8%	)	22	(	6.2%	)	72	(	20.3%	)	28	(	7.9%	)	50	(	14.1%	)	355	(	100.0%	)
Total	1	(	0.2%	)	3	(	0.6%	)	2	(	0.4%	)	38	(	7.7%	)	23	(	4.6%	)	41	(	8.3%	)	15	(	3.0%	)	124	(	25.0%	)	32	(	6.5%	)	97	(	19.6%	)	33	(	6.7%	)	87	(	17.5%	)	496	(	100.0%	)
Χ² = 18.5135 df = 11 p = .0704

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

Gender x Anticholinergic

Gender x RX_ANTICHOLINERGIC

Gender x OTC_ANTICHOLINERGIC

Gender x Smoking Status

	Smoking.Status
vo2_sex	Never				Former				Current				Total
1	62	(	44.0%	)	68	(	48.2%	)	11	(	7.8%	)	141	(	100.0%	)
2	204	(	57.5%	)	138	(	38.9%	)	13	(	3.7%	)	355	(	100.0%	)
Total	266	(	53.6%	)	206	(	41.5%	)	24	(	4.8%	)	496	(	100.0%	)
Χ² = 9.1257 df = 2 p = .0104

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

Smoking.Status x Weekly Alcohol Consumption Groups 1

	Weekly_alc.group_1
Smoking.Status	none				1-3				3-5				5-10				10+				Total
Never	120	(	45.1%	)	83	(	31.2%	)	23	(	8.6%	)	25	(	9.4%	)	15	(	5.6%	)	266	(	100.0%	)
Former	80	(	38.8%	)	61	(	29.6%	)	18	(	8.7%	)	24	(	11.7%	)	23	(	11.2%	)	206	(	100.0%	)
Current	7	(	29.2%	)	4	(	16.7%	)	5	(	20.8%	)	3	(	12.5%	)	5	(	20.8%	)	24	(	100.0%	)
Total	207	(	41.7%	)	148	(	29.8%	)	46	(	9.3%	)	52	(	10.5%	)	43	(	8.7%	)	496	(	100.0%	)
Χ² = 16.3033 df = 8 p = .0382

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

Smoking.Status x Weekly Alcohol Consumption Groups 2

	Weekly_alc.group_2
Smoking.Status	none				1-3				3+				Total
Never	120	(	45.1%	)	83	(	31.2%	)	63	(	23.7%	)	266	(	100.0%	)
Former	80	(	38.8%	)	61	(	29.6%	)	65	(	31.6%	)	206	(	100.0%	)
Current	7	(	29.2%	)	4	(	16.7%	)	13	(	54.2%	)	24	(	100.0%	)
Total	207	(	41.7%	)	148	(	29.8%	)	141	(	28.4%	)	496	(	100.0%	)
Χ² = 12.0223 df = 4 p = .0172

Generated by summarytools 0.9.9 (R version 3.6.1)
2021-06-18

# .

Model non-linear main effects

mod<- lm(hvlt_total_recall_tscore ~ 
           vo2_sex.factor+ 
           poly(vo2_age, 2)+ 
           poly(ses_year_education_2,2),
         data=df)


	Dependent variable:

	hvlt_total_recall_tscore

vo2_sex.factorFemale	5.2^*** (3.6, 6.9)
poly(vo2_age, 2)1	24.1^*** (7.8, 40.4)
poly(vo2_age, 2)2	-17.0^** (-33.2, -0.9)
poly(ses_year_education_2, 2)1	44.6^*** (28.3, 60.9)
poly(ses_year_education_2, 2)2	-17.5^** (-33.7, -1.3)
Constant	49.9^*** (48.6, 51.3)

Observations	496
R²	0.1
Adjusted R²	0.1
Residual Std. Error	8.2 (df = 490)
F Statistic	16.0^*** (df = 5; 490)

Note:	p<0.1; p<0.05; p<0.01

mod0<- lm(hvlt_total_recall_tscore ~ vo2_sex.factor,data=df)
mod1<- lm(hvlt_total_recall_tscore ~ Smoking.Status,data=df)
mod2<- lm(hvlt_total_recall_tscore ~ vo2_sex.factor+Smoking.Status,data=df)
mod3<- lm(hvlt_total_recall_tscore ~ Smoking.Status:vo2_sex,data=df)
mod4<- lm(hvlt_total_recall_tscore ~ vo2_sex.factor+ Smoking.Status:vo2_sex.factor,data=df)


	Dependent variable:

	hvlt_total_recall_tscore
	(1)	(2)	(3)	(4)	(5)

vo2_sex.factorFemale	4.53^***		4.63^***		4.28^***
	(2.85, 6.21)		(2.93, 6.32)		(1.83, 6.73)

Smoking.StatusFormer		0.73	1.17
		(-0.88, 2.33)	(-0.40, 2.75)

Smoking.StatusCurrent		-1.42	-0.37
		(-5.11, 2.28)	(-3.99, 3.24)

Smoking.StatusNever:vo2_sex				4.32^***
				(2.62, 6.01)

Smoking.StatusFormer:vo2_sex				5.04^***
				(3.25, 6.83)

Smoking.StatusCurrent:vo2_sex				3.72^***
				(0.97, 6.48)

vo2_sex.factorMale:Smoking.StatusFormer					0.32
					(-2.64, 3.29)

vo2_sex.factorFemale:Smoking.StatusFormer					1.54
					(-0.32, 3.40)

vo2_sex.factorMale:Smoking.StatusCurrent					1.46
					(-4.06, 6.98)

vo2_sex.factorFemale:Smoking.StatusCurrent					-2.07
					(-6.90, 2.76)

Constant	50.45^***	53.46^***	49.91^***	45.83^***	50.18^***
	(49.03, 51.87)	(52.40, 54.52)	(48.25, 51.57)	(42.84, 48.82)	(48.03, 52.32)


Observations	496	496	496	496	496
R²	0.05	0.003	0.06	0.06	0.06
Adjusted R²	0.05	-0.001	0.05	0.05	0.05
Residual Std. Error	8.61 (df = 494)	8.85 (df = 493)	8.61 (df = 492)	8.60 (df = 492)	8.61 (df = 490)
F Statistic	27.97^*** (df = 1; 494)	0.83 (df = 2; 493)	10.13^*** (df = 3; 492)	10.42^*** (df = 3; 492)	6.42^*** (df = 5; 490)

Note:	p<0.1; p<0.05; p<0.01

stargazer(model1, type = "html", ci = TRUE, digits = 1,  single.row = TRUE)


	Dependent variable:

	hvlt_total_recall_tscore

vo2_sex.factorFemale	5.2^*** (3.5, 6.8)
poly(ses_year_education_2, 2)1	43.8^*** (27.5, 60.2)
poly(ses_year_education_2, 2)2	-17.4^** (-33.5, -1.2)
poly(vo2_age, 2)1	24.2^*** (7.9, 40.5)
poly(vo2_age, 2)2	-17.6^** (-33.8, -1.5)
ANTICHOLINERGIC.factor1	2.0 (-0.4, 4.3)
Constant	49.8^*** (48.4, 51.2)

Observations	496
R²	0.1
Adjusted R²	0.1
Residual Std. Error	8.2 (df = 489)
F Statistic	13.8^*** (df = 6; 489)

Note:	p<0.1; p<0.05; p<0.01

model2<-lm(hvlt_total_recall_tscore ~
             poly(ses_year_education_2,2)+
             poly(vo2_age, 2)+
            Weekly_alc.score/vo2_sex.factor,
           data=df)

stargazer(model2, type = "html", ci = TRUE, digits = 1,  single.row = TRUE)


	Dependent variable:

	hvlt_total_recall_tscore

poly(ses_year_education_2, 2)1	39.0^*** (22.1, 55.8)
poly(ses_year_education_2, 2)2	-17.6^** (-34.3, -0.8)
poly(vo2_age, 2)1	20.4^** (3.6, 37.3)
poly(vo2_age, 2)2	-18.7^** (-35.5, -1.9)
Weekly_alc.score	-0.2^* (-0.4, 0.01)
Weekly_alc.score:vo2_sex.factorFemale	0.4^*** (0.1, 0.6)
Constant	53.6^*** (52.7, 54.5)

Observations	496
R²	0.1
Adjusted R²	0.1
Residual Std. Error	8.5 (df = 489)
F Statistic	7.5^*** (df = 6; 489)

Note:	p<0.1; p<0.05; p<0.01

Others:

LOGIT REGRESSION

https://stats.idre.ucla.edu/r/dae/logit-regression/

## 
## Call:
## glm(formula = poor_mem ~ vo2_sex.factor * Weekly_alc.score, family = "binomial", 
##     data = TMP)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -0.9470  -0.5802  -0.4581  -0.4406   2.1820  
## 
## Coefficients:
##                                       Estimate Std. Error z value Pr(>|z|)    
## (Intercept)                           -1.38252    0.26408  -5.235 1.65e-07 ***
## vo2_sex.factorFemale                  -0.90086    0.33569  -2.684  0.00728 ** 
## Weekly_alc.score                      -0.02358    0.03638  -0.648  0.51689    
## vo2_sex.factorFemale:Weekly_alc.score  0.10519    0.05057   2.080  0.03753 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 393.34  on 497  degrees of freedom
## Residual deviance: 384.07  on 494  degrees of freedom
## AIC: 392.07
## 
## Number of Fisher Scoring iterations: 4

MULTINOMIAL LOGISTIC REGRESSION

https://stats.idre.ucla.edu/r/dae/multinomial-logistic-regression/

Smoking.Status + Weekly_alc.score

## # weights:  15 (8 variable)
## initial  value 544.911695 
## iter  10 value 397.356781
## final  value 395.504125 
## converged

Weekly_alc.group_2 + Weekly_alc.score

## # weights:  15 (8 variable)
## initial  value 547.108920 
## iter  10 value 397.813510
## final  value 394.925001 
## converged

Current.Smoke.factor/prior_smoke_yrs

## # weights:  21 (12 variable)
## initial  value 544.911695 
## iter  10 value 404.428447
## iter  20 value 389.020615
## final  value 388.940276 
## converged

Heavy Drinking + EDU

## # weights:  12 (6 variable)
## initial  value 547.108920 
## iter  10 value 396.905342
## final  value 396.904863 
## converged

Sex * Weekly_alc.score

## # weights:  15 (8 variable)
## initial  value 547.108920 
## iter  10 value 393.033611
## final  value 392.560673 
## converged

Heavy Drinking * Weekly_alc.score

## # weights:  15 (8 variable)
## initial  value 547.108920 
## iter  10 value 392.743735
## final  value 392.623563 
## converged

IGNITE HHQ Variable Response Scales

Task: Look for basic correlations pertaining to smoking, alcohol usage, and general demographics using health history questionnaire.

Health History Questionnaire (HHQ)

Weekly Alcohol Consumption

Recode entries reported as...

1. Variable text entries

2. Subjective estimates outside of temporal scope

3. A passive range of units with...

4. Entries translated as dates

Alcohol Consumption Variables -

Smoking Demographics

Recode entries...

Missing Smoking Demographics Data

Smoking Demographics Variables

Write out .txt file...

Merge with processed/labeled RedCap database:

OTC/RX Anticholinergics

Unprocessed Baseline Medication Lists

Discrepancies between V02_data_beta, VA Classifiers, & custom fx to gather beta blocking ingredients.

.

Correlograms

Site Dispersion (trending/significant bolded)

Adjudication Outcomes

Demographics

Model non-linear main effects

LOGIT REGRESSION

MULTINOMIAL LOGISTIC REGRESSION