Per- and Polyfluoroalkyl Substances (PFAS) in Senior Care Facilities and Older Adult Residents

Abstract

Per- and polyfluoroalkyl substances (PFAS) are fluorinated organic compounds used in a variety of consumer products and industrial applications that persist in the environment, bioaccumulate in biological tissues, and can have adverse effects on human health, especially in vulnerable populations. In this study, we focused on PFAS exposures in seniors living in assisted living facilities. To investigate relationships between indoor, personal, and internal PFAS exposures, we analyzed 19 PFAS in matched samples of dust collected from the residents’ bedrooms, and wristbands and serum collected from the residents. The median ΣPFAS concentrations (the sum of all PFAS detected in the samples) measured in dust, wristbands, and serum were 120 ng/g, 0.05 ng/g, and 4.0 ng/mL, respectively. The most abundant compounds in serum were linear- and branched-perfluorooctane sulfonic acid (L-PFOS and B-PFOS, respectively) at medians of 1.7 ng/mL and 0.83 ng/mL, respectively, followed by the linear perfluorooctanoic acid (L-PFOA) found at a median concentration of 0.59 ng/mL. Overall, these three PFAS comprised 80% of the serum ΣPFAS concentrations. A similar pattern was observed in dust with L-PFOS and L-PFOA found as the most abundant PFAS (median concentrations of 13 and 7.8 ng/g, respectively), with the overall contribution of 50% to the ΣPFAS concentration. Only L-PFOA was found in wristbands at a median concentration of 0.02 ng/g. Significant correlations were found between the concentrations of several PFAS in dust and serum, and in dust and wristbands, suggesting that the indoor environment could be a significant contributor to the personal and internal PFAS exposures in seniors. Our findings demonstrate that residents of assisted living facilities are widely exposed to PFAS, with several PFAS found in blood of each study participant and in the assisted living environment.

Graphical Abstract

2.0. INTRODUCTION

Per- and polyfluoroalkyl substances (PFAS) are a class of fluorinated organic compounds that are used in a variety of industrial and commercial applications due to their strong carbon-fluorine bonds, chemical and thermal stability, and surfactant properties (Buck et al., 2011; Land et al., 2018). The historically widespread use of certain PFAS, such as perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA), led to the global spread and detection of these PFAS in the environment, wildlife, and humans (Armitage et al., 2006; Giesy and Kannan, 2001; Kärrman et al., 2010; Land et al., 2018; Olsen et al., 2005). PFOA and PFOS have been linked to developmental (Lau et al., 2003; Luebker et al., 2005b) and reproductive toxicity (Butenhoff et al., 2004; Lau et al., 2007; Luebker et al., 2005a), hepatotoxicity (DeWitt, 2015; Lau et al., 2007; Wang et al., 2015), and have been classified as carcinogens (Zahm et al., 2023). Due to their environmental persistence, bioaccumulation, and toxicity, production and use of these PFAS were restricted, starting in the early 2000s (EPA, 2000; EPA, 2024; UNEP). These changes led to the increased usage of their replacements, including shorter chain PFAS (Lu et al., 2019; Russell et al., 2013; Zheng et al., 2023). However, recent research has found evidence of acute toxicity from short-chain PFAS replacements to rodents and human mesenchymal stem cells (Chengelis et al., 2009; Iwai and Hoberman, 2014; Klaunig et al., 2015; Liu et al., 2020), showing that in some cases toxicity of these short-chain PFAS is higher than that of the long-chain counterparts (Wang et al., 2014).

Major exposure pathways to PFAS include diet (De Silva et al., 2021; Menzel et al., 2021; Sjogren et al., 2016; Tittlemier et al., 2007) and drinking water (Domingo and Nadal, 2019; Kotlarz et al., 2020; Rahman et al., 2014). However, recently the indoor environment has emerged as an important source of PFAS, and ingestion of indoor dust and inhalation of indoor air have also been shown as important PFAS exposure routes. PFAS have been detected in residential homes (Hall et al., 2020; Zheng et al., 2023) and public spaces serving sensitive populations (Wang et al., 2021; Zheng et al., 2020), including in dust, carpets, and nap mats at childcare facilities (Giovanoulis et al., 2019; Wu et al., 2020; Zheng et al., 2020), resulting in increased exposure in children through non-dietary ingestion and dermal absorption (Hauptman and Woolf, 2017; Poothong et al., 2020). Prenatal exposure to PFAS by expecting African American mothers has been associated with metabolic perturbations (Chang et al., 2022; Liang et al., 2023) and adverse birth outcomes, including lower birth weight and preterm births (Eick et al., 2023a; Eick et al., 2023b; Taibl et al., 2023a; Taibl et al., 2023b).

However, little research has been focused on another vulnerable population group, older adults. Elderly individuals spend 95% of their time indoors (Almeida-Silva et al., 2014), and about 1.3 million senior citizens (Sengupta et al., 2022) live in senior care facilities in the United States. Our previous research found high levels of several environmental contaminants in dust in U.S. assisted living communities (Arnold et al., 2018). These findings are of significant public health concern because elderly individuals can be more susceptible to environmental exposures and accumulation of chemicals due to weakened immune systems, slow metabolism, and reduced liver and kidney clearance efficiency (Choi et al., 2017; Jia et al., 2022). Despite this, limited research has focused on PFAS exposure in elderly populations. Existing research has found significant associations between PFAS exposure and adverse health effects in senior citizens (Salihovic et al., 2020; Wu et al., 2023). Some of these effects include immunotoxicity due to decreasing proteomic inflammatory markers (Salihovic et al., 2020) and increased risk of nonalcoholic fatty liver disease (Wu et al., 2023). This supports a need for a better understanding of PFAS exposures in this vulnerable population (Jia et al., 2022; Sjogren et al., 2016; Stubleski et al., 2017).

In this study, we have focused on examining PFAS exposures in older adults living in assisted living facilities in the United States. We analyzed 19 PFAS in matched samples of dust collected from assisted living residences in Indiana, United States, and in the wristband and blood serum collected from older adult residents to examine exposure patterns and determine the relationships between indoor, personal, and internal PFAS exposures to shed a light into possible exposure pathways and routes. To our best knowledge, this is the first study that has focused on PFAS exposures in assisted living facilities.

2.0. METHODS

2.1. Sample collection.

Serum, dust, and wristband samples (n = 40 sets of matched serum, dust, and wristbands; a total of 120 samples) were collected from three different assisted living facilities in Indiana, United States. All participants signed informed consent forms before participation and the research was approved by the Indiana University Institutional Review Board (Protocol #1909013530). All participants were at least 65 years old and had lived in the assisted living facility for at least 6 months at the time of recruitment.

Samples were collected between October 2021 – August 2022. Silicone wristbands were precleaned in the laboratory following established methods (Wang et al., 2019) distributed and worn for 72 hours by the participants. At the end of the 72-hour period, the wristbands were collected along with dust and blood samples from each resident. The dust samples were collected in each resident’s room in disposable pre-cleaned nylon socks using a previously developed method (Arnold et al., 2018). Blood samples were collected by venipuncture in vacutainer tubes. Samples were stored on ice and delivered to the laboratory at the end of each sampling day. The separation of the blood serum was performed immediately after being delivered to the laboratory. The serum samples were aliquoted intro polypropylene tubes and stored at −80°C until analysis. Wristbands and dust socks were stored wrapped in aluminum foil and sealed in separate zip lock bags at −20°C until analysis. Field blanks for dust, wristbands, and serum were collected during sample collection and treated the same as samples.

2.2. Sample analysis.

All samples were analyzed for 19 PFAS, including the C5-C12 perfluorocarboxylic acids (PFCAs) and C4, C6-C8, and C10 perfluoroalkanesulfonic acids (PFSAs), as well as hexafluoropropyleneoxide dimer acid (HFPO-DA), perfluorooctane sulfonamide (PFOSA), N-methyl perfluorooctane sulfonamide acetic acid (MeFOSAA), and N-ethyl perfluorooctane sulfonamidoacetic acid (EtFOSAA) using liquid chromatographic mass spectrometry (LC-MS/MS). The complete list of analytes and the details of the analytical methods and quality control measures (method detection limits [MDLs], blank levels, and National Institute of Standards and Technology (NIST) reference standards and matrix spike recoveries) are provided in the Supporting Information (Tables S1 – S3). Overall, the MDLs ranged from 0.07 to 1.0 ng/g for dust, from 0.003 to 0.07 ng/g for wristbands, and from 0.10 to 4.4 ng/mL for serum. The NIST recoveries were 70-130% for dust and serum reference standards. The majority of the matrix spike recoveries were within the 80-120% range (95%, 92%, and 100% of the analytes in dust, wristband, and serum, respectively), with a few exceptions (Table S3).

2.3. Data Analysis.

Descriptive statistics of the measured PFAS concentrations were characterized by selected percentiles (minimum, median, maximum) for each sample matrix. The reported results were blank corrected by subtracting the average blank concentrations from the sample concentrations prior to statistical analysis. The concentrations for the analytes in NIST reference standards were adjusted to an 80-120% recovery range (FDA, 2015; FDA, 2018). Percent contribution was determined as a ratio of the analyte’s median concentration to the ΣPFAS concentration (the sum of the concentrations of all PFAS detected in the samples).

Spearman correlation analysis was used to examine correlations among the analyte concentrations within and between matrices for the analytes that were detected in at least 50% of the samples. Concentrations that were below the MDLs were replaced by the $\text{MDL}/\sqrt{2}$ (Hornung and Reed, 1990) and were logarithmically (natural log) transformed for the analyses. All statistical analyses were performed using SigmaPlot 15.0 and R version 4.3.1.

3.0. RESULTS AND DISCUSSION

3.1. Concentrations.

The distribution of the PFAS concentrations measured in each matrix is presented in Table 1. Of the 19 PFAS targeted, 17 were detected in the analyzed samples with detection frequencies ranging from 3% to 100%. Perfluoropentanoic acid (PFPeA) and HFPO-DA were not detected in any samples and are not discussed further. Figure 1 shows the percent contribution of each PFAS to the ΣPFAS concentration (defined as a sum of all 17 PFAS detected) in each matrix.

Table 1:

Detection frequencies (DF, %), median (med), minimum (min), maximum (max), mean, and standard error (SE) for PFAS concentrations measured in dust, wristband, and serum samples collected from Indiana assisted living facilities and their residents. The percent contribution (contr., %) of each PFAS to the ΣPFAS concentrations (calculated based on medians). The non-detect (ND) values are values below the method detection limit (MDLs, Table S2).

	Dust (ng/g)					Wristband (ng/g)					Serum (ng/mL)
	DF	Med	Min – Max	Mean ± SE	Contr.	DF	Med	Min – Max	Mean ± SE	Contr.	DF	Med	Min – Max	Mean ± SE	Contr.
PFBS	69	5.0	ND – 340	44 ± 14	12	0	ND	-	-	-	3	ND	ND – 0.01	-	-
PFHxA	95	5.9	ND – 44	9.0 ± 1.5	14	31	ND	ND – 0.02	0.01 ± 0.001	-	0	ND	-	-	-
PFHxS	51	2.4	ND – 39	5.1 ± 1.3	5.8	3	ND	ND – 0.04	-	-	90	0.72	ND – 11	1.5 ± 0.36	18
PFHpA	95	2.8	ND – 30	4.8 ± 1.0	6.6	23	ND	ND – 0.03	0.01 ± 0.001	-	0	ND	-	-	-
PFHpS	10	ND	ND – 0.44	0.30 ± 0.02	-	0	ND	-	-	-	58	0.11	ND – 0.32	0.12 ± 0.02	2.8
L-PFOA	100	7.8	0.07 – 280	21 ± 7.7	19	74	0.02	ND – 0.12	0.03 ± 0.004	100	100	0.59	0.05 – 2.0	0.73 ± 0.08	15
B-PFOA	77	2.2	ND – 22	4.4 ± 0.92	5.2	36	ND	ND – 0.03	0.01 ± 0.001	-	3	ND	ND – 0.01	-	-
L-PFOS	92	13	ND – 140	20 ± 4.7	31	0	ND	-	-	-	100	1.7	0.07 – 6.7	2.1 ± 0.25	43
B-PFOS	44	ND	ND – 24	7.9 ± 1.2	-	0	ND	-	-	-	95	0.83	ND – 3.8	1.3 ± 0.15	21
PFOSA	31	ND	ND – 9.8	2.3 ± 0.51	-	18	ND	ND – 0.17	0.04 ± 0.01	-	0	ND	-	-	-
PFNA	90	0.96	ND – 28	3.2 ± 0.97	2.3	23	ND	ND – 0.03	0.01 ± 0.001	-	28	ND	ND – 1.5	0.46 ± 0.06	-
PFDA	59	2.0	ND – 130	11 ± 4.6	4.8	10	ND	ND – 0.06	0.04 ± 0.003	-	5	ND	ND – 0.21	0.14 ± 0.02	-
PFDS	44	ND	ND – 330	86 ± 13	-	0	ND	-	-	-	0	ND	-	-	-
PFUnDA	28	ND	ND – 33	4.8 ± 1.5	-	5	ND	ND – 0.20	0.11 ± 0.02	-	0	ND	-	-	-
PFDoDA	44	ND	ND – 96	11 ± 4.0	-	15	ND	ND – 0.03	0.02 ± 0.001	-	0	ND	-	-	-
MeFOSAA	18	ND	ND – 5.7	2.9 ± 0.26	-	3	ND	ND – 0.85	-	-	3	ND	ND – 0.70	-	-
EtFOSAA	46	ND	ND – 250	85 ± 12	-	26	ND	ND – 0.82	0.20 ± 0.04	-	0	ND	-	-	-
ΣPFAS		120	0.34 – 700	190 ± 32			0.05	ND – 2.0	0.12 ± 0.05			4.0	0.18 – 20	5.6 ± 0.73

Figure 1.

Contribution of each PFAS to the ΣPFAS concentrations in dust, wristbands, and serum (percent, calculated based on the analyte’s median concentration in each matrix). Only PFAS detected in at least 50% of samples were included.

Dust.

All 17 PFAS were detected in dust samples, with detection frequencies ranging from 10-100%. PFCAs were generally detected most frequently and were found in more than 50% of the samples, with the exception of perfluoroundecanoic acid (PFUnDA) and perfluoro-n-dodecanoic acid (PFDoDA) that were found in 28% and 44% of the samples, respectively. Among PFSAs, perfluoro-1-butanesulfonic acid (PFBS), perfluorohexanesulfonic acid (PFHxS), and linear PFOS (L-PFOS), were detected in 51-92% of the samples, and the rest of PFSAs were found less frequently.

The ΣPFAS concentrations in dust ranged from 0.34-700 ng/g (median 120 ng/g). The most abundant PFAS was L-PFOS measured at a median concentration of 13 ng/g with an overall contribution of 31% to the ΣPFAS concentration (Table 1 and Figure 1). Linear PFOA (L-PFOA) was the second most abundant PFAS found in dust samples (median 7.8 ng/g) and contributed 19% to the ΣPFAS concentrations. Branched isomers of PFOA and PFOS (B-PFOA and B-PFOS, respectively) were found less frequently and at lower concentrations. PFHxS was also measured at lower levels (median 2.4 ng/g) and contributed 6% to the ΣPFAS concentration. Interestingly, some of the short-chain PFAS were found more frequently and at levels exceeding or comparable to those of the legacy long-chain PFAS. For example, PFBS was found in 69% of the samples at a median level of 5.0 ng/g, while both perfluoro-n-hexanoic acid (PFHxA) and perfluoro-n-heptanoic acid (PFHpA) were found in 95% of the samples at median levels of 5.9 ng/g and 2.8 ng/g, respectively. Overall, PFCAs constituted the majority of the measured PFAS and contributed up to 52% to the ΣPFAS concentration; however, some of the PFSAs (such as PFBS and perfluoro-1-decanesulfonic acid [PFDS]) had the highest maximum concentrations reaching up to 340 ng/g. MeFOSAA and EtFOSAA were also detected in 18% and 46% of the samples at concentrations reaching up to 5.7 ng/g and 250 ng/g, respectively.

These results are consistent with the dust PFAS measurements collected in recent years. A study conducted in 2020 reported PFAS levels in the residential indoor dust collected from 81 homes across Indiana (Zheng et al., 2023). The PFOS levels found in the household dust samples were approximately half (median 5.9 ng/g) of the levels found in the assisted living facilities, while the PFOA levels in the residential dust samples (median 10 ng/g) were comparable to those measured here. Studies of residential homes in North Carolina, United States, detected median PFOS concentrations of 4.4-5.4 ng/g, which were also lower than the concentrations found in Indiana assisted living facilities, while the concentrations of PFOA found in the North Carolina samples (median 7.9-10 ng/g ) were comparable to those found in the present study (Hall et al., 2020). These findings suggest that there could be additional sources of PFOS in assisted living, contributing to higher exposures. This could be due to the use of different consumer products in the assisted living facility that contributed to higher PFOS concentrations (Herzke et al., 2012; Kotthoff et al., 2015). However, we were not able to identify these sources. In addition, similarly to this study, recent evidence shows the frequent detection of short-chain PFAS in the indoor environment, which suggests that the phase out of the longer-chain legacy PFAS and their replacement with shorter chain PFAS is now reflected in contaminant patterns observed in the environment (Yao et al., 2018; Zhang et al., 2020; Zheng et al., 2023). Other alternative PFAS, such as EtFOSAA, have also been measured at high concentrations in other studies (e.g., 422 ng/g in bedroom dust) (Winkens et al., 2018).

Wristbands.

Only a few PFAS, mostly PFCAs, were detected in 3-74% of the wristbands with most of them found in less than 40% of the samples. The most detected was L-PFOA found in 74% of the samples at a median concentration of 0.02 ng/g with a maximum concentration of 0.12 ng/g. B-PFOA was found in 36% of the samples at concentrations ranging from 0.003-0.03 ng/g. Interestingly, the short-chain PFHxA and PFHpA were also detected in 31% and 23% of the samples, respectively, at concentrations of up to 0.03 ng/g. Among PFSAs, only PFHxS was detected in a single sample at a concentration of 0.04 ng/g. MeFOSAA and EtFOSAA were detected at elevated concentrations in some samples reaching up to 0.85 ng/g but were found in just a few samples (detection frequency [DF] 3% and 26%, respectively). The more frequent detection of PFCAs compared to PFSAs in wristbands is similar to the trend found in dust samples analyzed here, suggesting that dust could be a source of these exposures captured by wristbands.

The low detection of specific PFAS in wristbands has also been found in previous research (Craig, 2020; Siebenaler and Cameron, 2016). Analysis of wristbands worn by children in a North Carolina study found low concentrations and detection of several ionic PFAS, such as PFCAs and PFSAs, with only perfluorononanoic acid (PFNA) and perfluoro-n-decanoic acid (PFDA) detected in more than 50% of the samples (Craig, 2020). Another study found high concentrations of PFBS and PFHxS on wristbands (2,500 ng/wristband and 62 ng/wristband, respectively), but at low detection frequencies (Siebenaler and Cameron, 2016). Higher detection and levels were found for PFHxS, PFOA, PFOS, PFNA, and PFDA in wristbands worn by off duty firefighters, including 0.4 ng/g for PFHxS, 1.4 ng/g for PFOA, and 0.7 ng/g for PFOS (Levasseur et al., 2022). The variations in concentrations and detection frequency of PFAS in wristbands could be partially due to silicone not being the best material for sampling of ionic PFAS, such PFCAs and PFSAs (Deng et al., 2023; Ortega-Calvo and Parsons, 2020). Frequent detection and higher concentrations of PFAS precursors and neutral PFAS have been reported previously for silicone wristband samples (Craig, 2020; Levasseur et al., 2022; Siebenaler and Cameron, 2016) and warrant analyzing an expanded suite of PFAS in wristbands. In addition, a recent study has reported higher detection frequencies of PFAS when using high density polyethylene (HDPE) or thermoplastic polyurethane (TPU) wristbands instead of silicone wristbands (Deng et al., 2023).

Serum.

Up to ten PFAS were detected in 3-100% of the serum samples. The ΣPFAS concentrations in serum ranged from 0.18-20 ng/mL. L-PFOS, B-PFOS, and L-PFOA were the most abundant PFAS detected in most of the samples. L-PFOS was measured at a median concentration of 1.7 ng/mL and B-PFOS was found at a median concentration of 0.83 ng/mL, and contributed 43% and 21%, respectively, to the serum ΣPFAS concentration, together comprising the majority of the ΣPFAS concentrations. L-PFOA (median 0.59 ng/mL) was found at lower levels compared to PFOS and contributed 15% to the ΣPFAS concentration, while B-PFOA was found only in a single sample (at a very low concentration of 0.01 ng/mL). PFHxS was detected in 90% of the samples at a median concentration of 0.72 ng/mL and contributed 18% to the ΣPFAS concentration. Interestingly, PFHxS concentrations were higher than the levels of L-PFOA, and it also had the highest maximum concentration out of all PFAS analyzed, reaching 11 ng/mL. Another frequently detected PFAS was the short-chain perfluoro-1-heptanesulfonic acid (PFHpS) found in 58% of the samples at a median concentration of 0.11 ng/mL and overall contribution of ~3% to the ΣPFAS concentration. PFNA was only found in 28% of the samples and at lower concentrations (0.21-1.5 ng/mL). Overall, PFSAs were detected more frequently and at higher concentrations in serum compared to PFCAs, a trend opposite to that observed in dust and wristband samples analyzed in this study. These differences may be due to different accumulation potentials and sources of PFAS in serum.

The levels of PFOS and PFOA found in this study were consistent with the findings in the 2017-2018 cycle of the National Health and Nutrition Examination Survey (NHANES) that reported median concentrations of 3.0 ng/mL and 1.3 ng/mL for these two compounds, respectively, in adult serum samples collected from the general U.S. population (NCHS, 2020). Studies from the elderly cohorts in Sweden reported detection of PFHxS, PFOS, PFOA at concentrations higher than those found in this study (medians of 2.1, 13, and 3.3, respectively) (Lind et al., 2014; Lind et al., 2022; Salihovic et al., 2020). Another study of the elderly in China found frequent detection of PFNA, PFDA, and PFUnDA in blood collected from this cohort at levels of 0.54, 0.16, and 0.29 ng/mL, respectively (Jia et al., 2022). These patterns are different from what we observe in the present study with no or low detection for compounds such as PFNA, PFDA, PFUnDa, and PFOSA. These variations could be due to differences in the regulations and timelines of phasing out legacy PFAS (Pan et al., 2018; Wang et al., 2017) and could be due to the differences in PFAS sources in the different countries (Kurwadkar et al., 2022). The levels for PFHxS, PFOS and PFOA measured here were comparable to those found in middle-aged adults from Indiana (0.7-1.5 ng/mL) (Zheng, 2023).

3.2. Concentration Correlations.

The correlations between the logarithmically transformed concentrations of PFAS detected in more than 50% of the samples within and across matrices were examined using Spearman correlation analysis and the results are presented in Figure 2 and Table S4.

Figure 2.

Spearman correlations for PFAS detected at over 50% detection frequency. Serum (S), dust (D), and wristband (W) PFAS are compared within and across matrices.

In serum, strong positive correlations were found among the concentrations of L-PFOS, B-PFOS, L-PFOA, PFHxS, and PFHpS (with the exception of PFHpS and L-PFOA), suggesting common sources for all of these PFAS (r = 0.32-0.82; p <0.05). Similarly, in dust highly significant positive correlations were found among the concentrations of almost all detected PFAS (r = 0.56-0.94; p < 0.05). Interestingly, PFBS did not correlate with any other PFAS in dust, suggesting that this short-chain PFSA could have a different source or have been recently introduced into the environment. PFBS also had the largest range and standard error and this variance in the data could also contribute to the lack of correlations with other PFAS in dust.

Across the matrices, strong positive correlations were found between most of the PFAS concentrations in dust with the L-PFOA in wristbands (r = 0.46-0.59, p < 0.05). The strongest relationships were found between the L-PFOA and B-PFOA levels in dust and L-PFOA in wristbands. In contrast, no significant correlations were found between the PFAS concentrations measured in wristbands and in serum. The strong correlations between the levels of PFAS on wristbands and those in dust suggest that the indoor environment is an important source of PFAS on the wristbands. On the other hand, the lack of significant correlations between PFAS in wristbands and those in serum suggests while wristbands could be a good proxy of PFAS exposures originating from the indoor environment, exposures captured by wristbands are not reflective of the internal body burden for the PFAS targeted in this study.

Moreover, concentrations for several PFAS in dust were strongly correlated with those in serum. PFHxS levels in serum were significantly and positively correlated with the dust levels of PFHxS, PFHpA, L-PFOA, B-PFOA, and PFDA (r = 0.31-0.44, p < 0.05). The serum PFHxS also had marginally significant correlations (0.05 < p < 0.1) with the levels of PFBS, PFHxA, and L-PFOS in dust. Strong relationships between PFHxS concentrations in serum and indoor dust have also been reported in previous studies (Siebenaler and Cameron, 2016; Zheng et al., 2023) and indicates that indoor dust ingestion could be an important exposure pathway for this PFAS. In addition to PFHxS, B-PFOS concentrations in serum marginally correlated with most PFAS measured in dust (r = 0.28-0.33; 0.05 < p < 0.1). Unlike B-PFOS, L-PFOS in serum was only marginally correlated with the PFHxS concentrations in dust (r = 0.30, p < 0.1). Previous research found a significant correlation between PFOS concentrations in dust and serum samples (Zheng et al., 2023), however our results indicate that the correlation might be driven by B-PFOS. Overall, these findings suggest that while indoor dust may be an important contributor to the bioaccumulation of certain PFAS in blood (such as PFHxS and B-PFOS), these exposures do not fully explain the total PFAS body burden and other exposure routes such as consumption of drinking water and diet could be important contributors. Our previous research suggests that dust ingestion and absorption explains only up to 7% of the PFAS body burden, while the remaining sources are unknown (Zheng et al., 2023).

3.3. Limitations.

This study had several limitations that narrowed the focus of the research. The study had a small sample size within a small geographic region of the country. This small sample size has limited the correlation analysis between the concentrations in different matrices. This study was also performed looking at only a limited list of PFAS and did not include PFAS precursors that may also be found in dust and wristbands. The research did not determine the contributions of PFAS from other exposure pathways such as drinking water and diet. These factors should be considered when interpreting the results of this research.

4.0. CONCLUSIONS

This is the first study that has focused on PFAS exposure in assisted living facilities and in older adults in the United States and explored the relationship between indoor exposures in this environment and personal and internal exposures in the residents. Our findings indicate that residents of assisted living facilities are widely exposed to PFAS, with several PFAS found in blood of each study participant. Significant correlations between certain dust and serum PFAS concentrations suggest that dust ingestion could be an important exposure route. However, they do not explain the whole extent of the internal body burden of PFAS and likely other exposure pathways such as drinking water and diet contribute. Overall, this study reports elevated levels of PFAS in U.S. assisted living and in older adults residing in these facilities and warrants further research in this vulnerable population to fully understand the extent and impact of the PFAS exposure.

Highlights.

• Several PFAS were detected in indoor dust, wristbands and serum blood

• PFOS and PFOA were most abundant in most of the samples

• Dust-wristband and dust-serum PFAS concentrations significantly correlated

ABBREVIATIONS

PFAS

Per- and Polyfluoroalkyl Substances

PFSA

perfluoroalkanesulfonic acids

PFCA

perfluorocarboxylic acids

PFPeA

Perfluoropentanoic acid

PFHxA

Perfluoro-n-hexanoic acid

PFHpA

Perfluoro-n-heptanoic acid

PFOA

Perfluorooctanoic acid

L-PFOA

Linear- Perfluorooctanoic acid

B-PFOA

Branchd- Perfluorooctanoic acid

PFNA

Perfluorononanoic acid

PFDA

Perfluoro-n-decanoic acid

PFUnDA

Perfluoroundecanoic acid

PFDoDA

Perfluoro-n-dodecanoic acid

PFBS

Perfluoro-1-butanesulfonic acid

PFHxS

Perfluorohexanesulfonic acid

PFHpS

Perfluoro-1-heptanesulfonic acid

PFOS

Perfluorooctanesulfonic acid

L-PFOS

Linear- Perfluorooctanesulfonic acid

B-PFOS

Branched- Perfluorooctanesulfonic acid

PFDS

Perfluoro-1-decanesulfonic acid

HFPO-DA

Hexafluoropropylene oxide dimer acid

PFOSA

Perfluorooctanesulfonamide

MeFOSAA

2-N-methyl-perfluorooctane sulfonamido acetic acid

EtFOSAA

2-N-ethyl-perfluorooctane sulfonamido acetic acid