-
Notifications
You must be signed in to change notification settings - Fork 0
/
analyze_questions.R
496 lines (394 loc) · 21.9 KB
/
analyze_questions.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
# analyze_questions.R
#
# Code to analyze associations between dust PFAS levels and exposure questionnaire
# responses using ANOVA.
library(psych)
library(dplyr)
library(ggplot2)
library(tidyr)
library(lme4)
library(car)
library(FSA)
# load pre-processed individual-level data with non-detects as DL / sqrt(2)
df = read.csv('output/data_all_sites_questionnaire.csv', stringsAsFactors=TRUE, na.strings= c('NA', ''))
PFAS = c('PFOA', 'PFOS', 'PFNA', 'PFDA', 'PFUnA', 'PFHxS', 'MeFOSAA')
# make unique household id column
df$hid = paste(df$site, df$householdid, sep="")
length(unique(df$hid)) # 114 households
# convert concs to log
PFAS = c('PFOA', 'PFOS', 'PFNA', 'PFDA', 'PFUnA', 'PFHxS', 'MeFOSAA')
for (substance in PFAS) {
df[paste('log_', substance, '_dust', sep="")] = log(df[paste(substance, '_dust_ng.g', sep="")])
df[paste('log_', substance, '_serum', sep="")] = log(df[paste(substance, '_serum_ng.ml', sep="")])
}
# calculate sum of PFAS in serum + dust
df$sum_PFAS_serum_ng.ml <- df$PFOA_serum_ng.ml + df$PFOS_serum_ng.ml + df$PFHxS_serum_ng.ml + df$PFDA_serum_ng.ml + df$PFNA_serum_ng.ml +
df$MeFOSAA_serum_ng.ml + df$PFUnA_serum_ng.ml
df$sum_PFAS_dust_ng.g <- df$PFOA_dust_ng.g + df$PFOS_dust_ng.g + df$PFHxS_dust_ng.g + df$PFDA_dust_ng.g + df$PFNA_dust_ng.g +
df$MeFOSAA_dust_ng.g + df$PFUnA_dust_ng.g
df$log_sum_PFAS_serum = log(df$sum_PFAS_serum_ng.ml)
df$log_sum_PFAS_dust = log(df$sum_PFAS_dust_ng.g)
df$log_dust_load <- log(df$dust_load_g_m2)
# calculate dust load variables
for (substance in PFAS) {
df[paste('log_', substance, '_dust_load', sep="")] = log(df[paste(substance, '_dust_ng.g', sep="")] * df$dust_load_g_m2)
}
df["log_sum_PFAS_dust_load"] = log(df['sum_PFAS_dust_ng.g'] * df$dust_load_g_m2)
###########
### basic demographics for the dataset
###########
summary(df$sex) # Female: 115, Male 87
df_complete = df[complete.cases(df$age),] # remove 1 age non-response
nrow(df_complete[df_complete$age < 18,]) # age < 18 : 17
nrow(df_complete[(df_complete$age >= 18) & (df_complete$age < 50),]) # age >= 18 but < 50: 57
nrow(df_complete[df_complete$age >= 50,]) # age >= 50: 127
# N for white, hispanic
nrow(df[(df$AQ1_Ethnicity %in% 'True') & (df$AQ2_White %in% 'True'),]) # Adults 3
nrow(df[(df$CQ2_Ethnicity%in%'True') & (df$AQ2_White%in%'True'),]) # Children 3
# N for white, non-hispanic
nrow(df[(df$AQ1_Ethnicity %in% 'False') & (df$AQ2_White %in% 'True'),]) # Adults 162
nrow(df[(df$CQ2_Ethnicity%in%'False') & (df$AQ2_White%in%'True'),]) # Children 13
# N for not white
nrow(df[df$AQ2_White %in% 'False',]) # Adults + children 17
#range of participants per house
max(table(df$hid)) # max 6
min(table(df$hid)) # min 1
###########
### analyze mean HH age (and max HH age)
###########
# aggregate within households
dfhh_age = df[ , grepl( "dust|hid|site|age" , names( df ) ) ]
dfhh_age = as.data.frame(dfhh_age %>%
group_by(hid, site) %>%
summarise(across(everything(), list(mean=mean,max=max))))
nrow(dfhh_age) # 114
plot(log_PFOS_dust_mean ~ age_mean, data=dfhh_age)
plot(log_PFOA_dust_mean ~ age_mean, data=dfhh_age)
plot(log_PFHxS_dust_mean ~ age_mean, data=dfhh_age)
plot(log_PFNA_dust_mean ~ age_mean, data=dfhh_age)
plot(log_PFDA_dust_mean ~ age_mean, data=dfhh_age)
plot(log_MeFOSAA_dust_mean ~ age_mean, data=dfhh_age)
plot(log_PFUnA_dust_mean ~ age_mean, data=dfhh_age)
plot(log_sum_PFAS_dust_mean ~ age_mean, data=dfhh_age)
Anova(lm(log_PFOS_dust_mean ~ age_mean + site, data=dfhh_age)) # sig
Anova(lm(log_PFOA_dust_mean ~ age_mean + site, data=dfhh_age)) # sig
Anova(lm(log_PFHxS_dust_mean ~ age_mean + site, data=dfhh_age)) # sig
Anova(lm(log_PFNA_dust_mean ~ age_mean + site, data=dfhh_age)) # sig
Anova(lm(log_PFDA_dust_mean ~ age_mean + site, data=dfhh_age)) # sig
Anova(lm(log_MeFOSAA_dust_mean ~ age_mean + site, data=dfhh_age)) # sig
Anova(lm(log_PFUnA_dust_mean ~ age_mean + site, data=dfhh_age)) # sig
Anova(lm(log_sum_PFAS_dust_mean ~ age_mean + site, data=dfhh_age)) # sig
Anova(lm(log_PFOS_dust_mean ~ age_max + site, data=dfhh_age)) # sig
Anova(lm(log_PFOA_dust_mean ~ age_max + site, data=dfhh_age)) # sig
Anova(lm(log_PFHxS_dust_mean ~ age_max + site, data=dfhh_age)) # sig
Anova(lm(log_PFNA_dust_mean ~ age_max + site, data=dfhh_age)) # sig
Anova(lm(log_PFDA_dust_mean ~ age_max + site, data=dfhh_age)) # sig
Anova(lm(log_MeFOSAA_dust_mean ~ age_max + site, data=dfhh_age)) # sig
Anova(lm(log_PFUnA_dust_mean ~ age_max + site, data=dfhh_age)) # sig
Anova(lm(log_sum_PFAS_dust_mean ~ age_max + site, data=dfhh_age)) # sig
###########
### analyze years at current address
###########
# aggregate within households
years_at_home_years = df$AQ3_Years
years_at_home_years[is.na(years_at_home_years)] = 0
years_at_home_months = df$AQ3_Months/12
years_at_home_months[is.na(years_at_home_months)] = 0
df$years_at_home = years_at_home_years + years_at_home_months
dfhh_years_at_home= df[ , grepl( "dust|hid|site|years|age" , names( df ) ) ]
dfhh_years_at_home = as.data.frame(dfhh_years_at_home %>%
group_by(hid, site) %>%
summarise(across(everything(), mean)))
nrow(dfhh_years_at_home) # 114
# PFOS
mod = lm(log_PFOS_dust ~ years_at_home + site, data=dfhh_years_at_home)
Anova(mod) # ANOVA sig +;
hist(resid(mod))
plot(log_PFOS_dust ~ years_at_home, data=dfhh_years_at_home)
# PFOA
mod = lm(log_PFOA_dust ~ years_at_home + site, data=dfhh_years_at_home)
Anova(mod) # ANOVA sig +
hist(resid(mod))
plot(log_PFOA_dust ~ years_at_home, data=dfhh_years_at_home)
# PFHxS
mod = lm(log_PFHxS_dust ~ years_at_home + site, data=dfhh_years_at_home)
Anova(mod) # ANOVA sig +
hist(resid(mod))
plot(log_PFHxS_dust ~ years_at_home, data=dfhh_years_at_home)
# PFNA
mod = lm(log_PFNA_dust ~ years_at_home + site, data=dfhh_years_at_home)
Anova(mod) # ANOVA sig +
hist(resid(mod))
plot(log_PFNA_dust ~ years_at_home, data=dfhh_years_at_home)
# PFDA
mod = lm(log_PFDA_dust ~ years_at_home + site, data=dfhh_years_at_home)
Anova(mod) # ANOVA SIG +
hist(resid(mod))
plot(log_PFDA_dust ~ years_at_home, data=dfhh_years_at_home)
# MeFOSAA
mod = lm(log_MeFOSAA_dust ~ years_at_home + site, data=dfhh_years_at_home)
Anova(mod) # ANOVA sig +
hist(resid(mod))
plot(log_MeFOSAA_dust ~ years_at_home, data=dfhh_years_at_home)
# PFUnA
mod = lm(log_PFUnA_dust ~ years_at_home + site, data=dfhh_years_at_home)
Anova(mod) # ANOVA NOT sig
hist(resid(mod))
plot(log_PFUnA_dust ~ years_at_home, data=dfhh_years_at_home)
# Sum of PFAS
mod = lm(log_sum_PFAS_dust ~ years_at_home + site, data=dfhh_years_at_home)
Anova(mod) # ANOVA SIG
hist(resid(mod))
# age vs years at home (individual-level)
plot(years_at_home ~ age, data=df)
cor.test(df$years_at_home, df$age, method=c('pearson')) # SIG r=0.46
# age vs years at home (household-level)
cor.test(dfhh_years_at_home$years_at_home, dfhh_years_at_home$age, method=c('pearson')) # SIG r=0.51
# does years at home relate to dust load in the house
Anova(lm(log_dust_load ~ years_at_home + site, data=dfhh_years_at_home)) # no effect on dust load
### NOTES
# years lived at current home is sig. positively associated with
# dust levels for 6 of 7 PFAS (All but PFUnA)
# also associated for sum(PFAS)
# May correlate with age of home or age of elements inside the home
###########
### analyze cleaning frequency
###########
summary(df$AQ14_Cleaning)
df$cleaning_bin = ifelse(df$AQ14_Cleaning == "Three times per week or more", 1, 0)
df$cleaning_bin[is.na(df$cleaning_bin)] = 0
dfhh_clean = df[ , grepl( "dust|hid|site|age|cleaning_bin" , names( df ) ) ]
dfhh_clean = as.data.frame(dfhh_clean %>%
group_by(hid, site) %>%
summarise(across(everything(), list(mean=mean,max=max))))
dfhh_clean$cleaning_bin_max = factor(dfhh_clean$cleaning_bin_max, labels=c('Low/Medium', 'High'))
nrow(dfhh_clean) # 114
summary(dfhh_clean$cleaning_bin_max) # 25 High frequency, 79 Medium/Low frequency
Anova(lm(log_PFOS_dust_mean ~ cleaning_bin_max + site, data=dfhh_clean)) # SIG (p=0.007)
Anova(lm(log_PFOA_dust_mean ~ cleaning_bin_max + site, data=dfhh_clean)) # not sig (p=0.12)
Anova(lm(log_PFHxS_dust_mean ~ cleaning_bin_max + site, data=dfhh_clean)) # SIG (p=0.01)
Anova(lm(log_PFNA_dust_mean ~ cleaning_bin_max + site, data=dfhh_clean)) # not sig (p=0.48)
Anova(lm(log_PFDA_dust_mean ~ cleaning_bin_max + site, data=dfhh_clean)) # not sig (p=0.58)
Anova(lm(log_MeFOSAA_dust_mean ~ cleaning_bin_max + site, data=dfhh_clean)) # SIG (p=0.001)
Anova(lm(log_PFUnA_dust_mean ~ cleaning_bin_max + site, data=dfhh_clean)) # not sig (p=0.97)
Anova(lm(log_sum_PFAS_dust_mean ~ cleaning_bin_max + site, data=dfhh_clean)) # SIG (p=0.04)
# does cleaning frequency affect dust load in the house
Anova(lm(log_dust_load_max ~ cleaning_bin_max + site, data=dfhh_clean)) # no effect on dust load
### NOTES
# 3 of 7 PFAS (PFOS, PFHxS, MeFOSAA) are associated with household cleaning frequency
# Also sum(PFAS) is associated with cleaning frequency
# cleaning freq vs age
plot(age_mean ~ factor(cleaning_bin_max), data=dfhh_clean)
summary(aov(age_mean ~ cleaning_bin_max + site, data=dfhh_clean)) # households with older mean age cleaned less
# cleaning frequency effect sizes
(exp(3.07385-0.96652) - exp(3.07385)) / exp(3.07385) * 100 # PFOS
(exp(2.1734-0.9706) - exp(2.1734)) / exp(2.1734) * 100 # PFHxS
(exp(1.77760-1.08013) - exp(1.77760)) / exp(1.77760) * 100 # MeFOSAA
(exp(4.46286-0.66427) - exp(4.46286)) / exp(4.46286) * 100 # SUM PFAS
###########
### analyze stain resistant products
###########
summary(df$AQ15_StainResist)
df$stain_bin = ifelse(df$AQ15_StainResist == "Never", 0, 1)
df$stain_bin[is.na(df$stain_bin)] = 0
dfhh_stain = df[ , grepl( "dust|hid|site|stain_bin|age" , names( df ) ) ]
dfhh_stain = as.data.frame(dfhh_stain %>%
group_by(hid, site) %>%
summarise(across(everything(), list(mean=mean,max=max))))
dfhh_stain$stain_bin_max = factor(dfhh_stain$stain_bin_max, labels=c('No','Yes'))
nrow(dfhh_stain) # 114
summary(dfhh_stain$stain_bin_max) # yes = 15, no = 99
# ANOVAs
Anova(lm(log_PFOS_dust_mean ~ stain_bin_max + site, data=dfhh_stain))
Anova(lm(log_PFOA_dust_mean ~ stain_bin_max + site, data=dfhh_stain))
Anova(lm(log_PFHxS_dust_mean ~ stain_bin_max + site, data=dfhh_stain))
Anova(lm(log_PFNA_dust_mean ~ stain_bin_max + site, data=dfhh_stain)) # SIG (p=0.002)
Anova(lm(log_PFDA_dust_mean ~ stain_bin_max + site, data=dfhh_stain)) # SIG (p=0.011)
Anova(lm(log_MeFOSAA_dust_mean ~ stain_bin_max + site, data=dfhh_stain))
Anova(lm(log_PFUnA_dust_mean ~ stain_bin_max + site, data=dfhh_stain)) # SIG (p<0.001)
Anova(lm(log_sum_PFAS_dust_mean ~ stain_bin_max + site, data=dfhh_stain))
# age vs stain resistant product use
summary(aov(age_mean ~ stain_bin_max, data=dfhh_stain)) # no rel. between age and SR product use
# stain resistant product use effect sizes
(exp(1.21980+0.85496) - exp(1.21980)) / exp(1.21980) * 100 # PFNA
(exp(1.37293+0.67148 ) - exp(1.37293)) / exp(1.37293) * 100 # PFDA
(exp(0.57252+0.99208 ) - exp(0.57252)) / exp(0.57252) * 100 # PFUnA
### NOTES
# 3 of 7 PFAS (PFNA, PFDA, PFUnA) are associated with stain resistant product use)
# sum of PFAS not associated
###########
### analyze flooring
###########
# create flooring household df
dfhh_floors = df[ , grepl( "dust|hid|site|Floor|age" , names( df ) ) ]
dfhh_floors = as.data.frame(dfhh_floors %>%
group_by(hid, site) %>%
summarise(across(everything(), list(mean=mean,first=first))))
nrow(dfhh_floors) # 114
# taking first listed per hh to be the flooring
dfhh_floors$AQ16_FloorLR = factor(dfhh_floors$AQ16_FloorLR_first)
dfhh_floors$AQ17_FloorK = factor(dfhh_floors$AQ17_FloorK_first)
dfhh_floors$AQ18_FloorB = factor(dfhh_floors$AQ18_FloorB_first)
summary(dfhh_floors$AQ16_FloorLR)
summary(dfhh_floors$AQ17_FloorK)
summary(dfhh_floors$AQ18_FloorB)
# aggregate to hard vs soft flooring (except kitchen since no soft type there)
dfhh_floors$living_room <- recode(dfhh_floors$AQ16_FloorLR, "c('Hardwood', 'Laminate',
'Tile', 'Vinyl')='hard';c('Carpet') = 'soft'")
dfhh_floors$kitchen <- dfhh_floors$AQ17_FloorK_first # kitchen has no soft type
dfhh_floors$bedroom <- recode(dfhh_floors$AQ18_FloorB_first, "c('Hardwood', 'Laminate',
'Tile', 'Vinyl')='hard';c('Carpet') = 'soft'")
dfhh_floors[dfhh_floors$living_room == "Other","living_room"] <- NA
dfhh_floors[dfhh_floors$bedroom == "Other", "bedroom"] <- NA
dfhh_floors$living_room <- droplevels(dfhh_floors$living_room)
dfhh_floors$bedroom <- droplevels(dfhh_floors$bedroom)
summary(dfhh_floors$living_room) # hard = 53, soft = 60, NA = 1
summary(dfhh_floors$bedroom) # hard = 33, soft = 79, NA = 2
### Living Room
Anova(lm(log_PFOS_dust_mean ~ living_room + site, data=dfhh_floors))
Anova(lm(log_PFOA_dust_mean ~ living_room + site, data=dfhh_floors))
Anova(lm(log_PFHxS_dust_mean ~ living_room + site, data=dfhh_floors))
Anova(lm(log_PFNA_dust_mean ~ living_room + site, data=dfhh_floors))
Anova(lm(log_PFDA_dust_mean ~ living_room + site, data=dfhh_floors)) # SIG Soft > Hard
TukeyHSD(aov(log_PFDA_dust_mean ~ living_room + site, data=dfhh_floors))
Anova(lm(log_MeFOSAA_dust_mean ~ living_room + site, data=dfhh_floors))
Anova(lm(log_PFUnA_dust_mean ~ living_room + site, data=dfhh_floors)) # SIG soft > Hard
TukeyHSD(aov(log_PFUnA_dust_mean ~ living_room + site, data=dfhh_floors))
Anova(lm(log_sum_PFAS_dust_mean ~ living_room + site, data=dfhh_floors))
# LR vs age
summary(aov(age_mean ~ living_room, data=dfhh_floors)) # SIG
TukeyHSD(aov(age_mean ~ living_room, data=dfhh_floors))# older household more likely to have carpet than Vinyl
plot(age_mean ~ AQ16_FloorLR, data=dfhh_floor_LR)
# effect sizes
(exp(1.19458+0.55) - exp(1.19458)) / exp(1.19458) * 100 # PFDA
(exp(0.489912+0.423566 ) - exp(0.489912)) / exp(0.489912) * 100 # PFDA
### Kitchen - not examined because only 2 households had soft type flooring
### Bedroom
Anova(lm(log_PFOS_dust_mean ~ bedroom + site, data=dfhh_floors))
Anova(lm(log_PFOA_dust_mean ~ bedroom + site, data=dfhh_floors))
Anova(lm(log_PFHxS_dust_mean ~ bedroom + site, data=dfhh_floors))
Anova(lm(log_PFNA_dust_mean ~ bedroom + site, data=dfhh_floors))
Anova(lm(log_PFDA_dust_mean ~ bedroom + site, data=dfhh_floors))
Anova(lm(log_MeFOSAA_dust_mean ~ bedroom + site, data=dfhh_floors))
Anova(lm(log_PFUnA_dust_mean ~ bedroom + site, data=dfhh_floors))
Anova(lm(log_sum_PFAS_dust_mean ~ bedroom + site, data=dfhh_floors))
# bedroom vs age
summary(aov(age_mean ~ AQ18_FloorB, data=dfhh_floor_B)) # not sig
plot(age_mean ~ AQ18_FloorB, data=dfhh_floor_B)
# does floor type relate to dust load in the house
Anova(lm(log_dust_load_mean ~ bedroom + site, data=dfhh_floors)) # SIG effect on dust load
Anova(lm(log_dust_load_mean ~ living_room + site, data=dfhh_floors)) # SIG effect on dust load
plot(log_dust_load_mean ~ bedroom, data=dfhh_floors) # SIG effect on dust load
plot(log_dust_load_mean ~ living_room, data=dfhh_floors) # SIG effect on dust load
###########
### analyze occupational exposure
###########
df$work_bin = ifelse(df$AQ28_None == 'False', 1, 0)
df$work_bin[is.na(df$work_bin)] = 0
df$work_bin[df$AQ28_DontKnow == 'True'] = 0
dfhh_work = df[ , grepl( "dust|hid|site|work_bin|age" , names( df ) ) ]
dfhh_work = as.data.frame(dfhh_work %>%
group_by(hid, site) %>%
summarise(across(everything(), list(mean=mean,max=max))))
nrow(dfhh_work) # 114
dfhh_work$work_bin_max = factor(dfhh_work$work_bin_max, labels=c('no', 'yes')) # indicates if anyone in hh has worked in those industries
summary(dfhh_work$work_bin_max) # yes = 16, no = 98
plot(log_PFOS_dust_mean ~ work_bin_max, data=dfhh_work)
plot(log_PFOA_dust_mean ~ work_bin_max, data=dfhh_work)
plot(log_PFHxS_dust_mean ~ work_bin_max, data=dfhh_work)
plot(log_PFNA_dust_mean ~ work_bin_max, data=dfhh_work)
plot(log_PFDA_dust_mean ~ work_bin_max, data=dfhh_work)
plot(log_MeFOSAA_dust_mean ~ work_bin_max, data=dfhh_work)
plot(log_PFUnA_dust_mean ~ work_bin_max, data=dfhh_work)
Anova(lm(log_PFOS_dust_mean ~ work_bin_max + site, data=dfhh_work)) # not sig
Anova(lm(log_PFOA_dust_mean ~ work_bin_max + site, data=dfhh_work)) # not sig
Anova(lm(log_PFHxS_dust_mean ~ work_bin_max + site, data=dfhh_work)) # not sig
Anova(lm(log_PFNA_dust_mean ~ work_bin_max + site, data=dfhh_work)) # not sig
Anova(lm(log_PFDA_dust_mean ~ work_bin_max + site, data=dfhh_work)) # not sig
Anova(lm(log_MeFOSAA_dust_mean ~ work_bin_max + site, data=dfhh_work)) # not sig
Anova(lm(log_PFUnA_dust_mean ~ work_bin_max + site, data=dfhh_work)) # not sig
Anova(lm(log_sum_PFAS_dust_mean ~ work_bin_max + site, data=dfhh_work)) # not sig
### NOTES
# No significant trends between PFAS levels in dust and someone in HH
# working in industries with possible exposure.
# associated with age? no
summary(aov(age_mean ~ work_bin_max, data=dfhh_work))
###########
### raceðnicity
###########
# define households with at least 1 non-white or white, Hispanic participant
df$white_nonhispanic = ifelse((df$AQ2_White %in% 'True') &
!((df$AQ1_Ethnicity %in% 'True') | (df$CQ2_Ethnicity %in% 'True')), 1,0)
dfhh_race = df[ , grepl( "dust|hid|site|white_nonhispanic|age" , names( df ) ) ]
dfhh_race = as.data.frame(dfhh_race %>%
group_by(hid, site) %>%
summarise(across(everything(), list(mean=mean,all=min))))
nrow(dfhh_race) # n = 114
dfhh_race$white_nonhispanic_all = factor(dfhh_race$white_nonhispanic_all, labels=c('no', 'yes'))
summary(dfhh_race$white_nonhispanic_all) # yes (all participants white, non-hispanic) = 103, no = 13
Anova(lm(log_PFOS_dust_mean ~ white_nonhispanic_all + site, data=dfhh_race)) # not sig
Anova(lm(log_PFOA_dust_mean ~ white_nonhispanic_all + site, data=dfhh_race)) # not sig
Anova(lm(log_PFHxS_dust_mean ~ white_nonhispanic_all + site, data=dfhh_race)) # not sig
Anova(lm(log_PFNA_dust_mean ~ white_nonhispanic_all + site, data=dfhh_race)) # not sig
Anova(lm(log_PFDA_dust_mean ~ white_nonhispanic_all + site, data=dfhh_race)) # not sig
Anova(lm(log_MeFOSAA_dust_mean ~ white_nonhispanic_all + site, data=dfhh_race)) # not sig
Anova(lm(log_PFUnA_dust_mean ~ white_nonhispanic_all + site, data=dfhh_race)) # not sig
Anova(lm(log_sum_PFAS_dust_mean ~ white_nonhispanic_all + site, data=dfhh_race)) # not sig
### NOTES
# No significant trends between PFAS levels in dust and someone in HH
# identifying as something other than white and non-Hispanic
# But there were only 13/114 households with someone identifying as
# anything other than white and non-Hispanic.
####################
###########
### Cleaning frequency manuscript plots
###########
# reform dataframe from wide to long
log_dust_cols <- c('log_PFOA_dust_mean', 'log_PFOS_dust_mean', 'log_PFHxS_dust_mean',
'log_PFNA_dust_mean', 'log_PFDA_dust_mean', 'log_PFUnA_dust_mean',
'log_MeFOSAA_dust_mean', 'log_sum_PFAS_dust_mean')
df_clean_long <- melt(dfhh_clean, id.vars=c('site', 'cleaning_bin_max'), measure.vars=log_dust_cols,
variable.name='chemical', value.name='dust_ng.g')
chemical_labels <- c('PFOA', 'bold(PFOS)', 'bold(PFHxS)', 'PFNA', 'PFDA', 'PFUnA', 'bold(MeFOSAA)', 'bold("Sum PFAS")')
df_clean_long$chemical_labeled <- factor(df_clean_long$chemical, labels=chemical_labels)
df_clean_long$cleaning_level = factor(df_clean_long$cleaning_bin_max, labels=c('Low/Med', 'High'))
plot_cleaning <- ggplot(df_clean_long[complete.cases(df_clean_long$cleaning_level),],
aes(cleaning_level, dust_ng.g, fill=cleaning_level)) + facet_wrap(~chemical_labeled,nrow=1, labeller = label_parsed) +
geom_boxplot() + scale_fill_manual(values=c("grey90", "grey60")) +#ylim(0,200)
xlab("Cleaning frequency") + ylab("Concentration in dust (log ng/g)") +
theme_classic() + theme(axis.title = element_text(size = 14),
axis.text = element_text(size = 12),
plot.title = element_text(size = 18, hjust = 0.5),
legend.position = "none",
panel.background = element_rect(fill = NA, color = "black"),
strip.text.x = element_text(size = 14))
###########
### SR product use manuscript plots
###########
log_dust_cols <- c('log_PFOA_dust_mean', 'log_PFOS_dust_mean', 'log_PFHxS_dust_mean',
'log_PFNA_dust_mean', 'log_PFDA_dust_mean', 'log_PFUnA_dust_mean',
'log_MeFOSAA_dust_mean', 'log_sum_PFAS_dust_mean')
df_stain_long <- melt(dfhh_stain, id.vars=c('site', 'stain_bin_max'), measure.vars=log_dust_cols,
variable.name='chemical', value.name='dust_ng.g')
chemical_labels <- c('PFOA', 'PFOS', 'PFHxS', 'bold(PFNA)', 'bold(PFDA)', 'bold(PFUnA)', 'MeFOSAA', '"Sum PFAS"')
df_stain_long$chemical_labeled <- factor(df_stain_long$chemical, labels=chemical_labels)
label_shade_df <- data.frame(
var = log_dust_cols,
var_color = c("white", "white", "white", "grey50", "grey50", "grey50", "white", "white")
)
plot_stain <- ggplot(df_stain_long[complete.cases(df_stain_long$stain_bin_max),]) +
facet_wrap(~chemical_labeled,nrow=1, labeller = label_parsed) +
geom_boxplot(aes(stain_bin_max, dust_ng.g, fill=stain_bin_max)) + scale_fill_manual(values=c("grey90", "grey60")) +#ylim(0,200)
xlab("Use of stain-resistant products") + ylab("Concentration in dust (log ng/g)") +
theme_classic() + theme(axis.title = element_text(size = 14),
axis.text = element_text(size = 12),
plot.title = element_text(size = 18, hjust = 0.5),
legend.position = "none",
panel.background = element_rect(fill = NA, color = "black"),
strip.text.x = element_text(size = 14))
###########
### combine cleaning frequency and SR product use plots
###########
output = plot_grid(plot_cleaning, plot_stain,align='h',nrow=2, labels=c("a)", "b)"),scale=0.93)# +
#draw_label("Concentration in dust (log ng/g)", x= 0.03, y=0.5, vjust= 1.5, angle=90)
ggsave("output/figures/Figure_A2.jpeg", output, width = 12, height = 7, units = "in", dpi = 600)