Introduction

Evidence before this study

Links between environmental chemicals and human health in the American adults including hypertension, cardiovascular disease, food allergy, oral health, emotional support, and cognitive function have emerged (Shiue 2015a, b, c, d, 2014a, b, 2013a, b, c), but the effects on sleep health, such as time to fall asleep as one of the indicators, were relatively less studied. Epidemiological studies have shown that the prevalence of sleep disturbances lies between 20 and 30% and increases with age, particularly in female (Zeitlhofer et al. 2000). Among the American adults, in studying risk contributors for sleep health it was observed that vitamin D level, a hormone that interacts with intranuclear receptors to effect transcriptional changes in many cell types including those in gut, bone, breast, prostate, brain, skeletal muscle, and the immune system (McCarty et al. 2014), could also play an important role in time to fall asleep at bedtime (Shiue 2013d).

Study aim

However, it is unclear whether environmental chemicals in addition to sun radiation could influence sleep patterns as well. Following this context, therefore, the aim of the present study was to investigate the relationships of different sets of environmental chemicals and some common sleep troubles in a national and population-based setting using an independent dataset.

Methods

Study sample

As described online (more details via http://www.cdc.gov/nchs/nhanes.htm), the United States National Health and Nutrition Examination Surveys (NHANES) has been a national, population-based, multi-year, cross-sectional study since the 1980s. Study samples are representative of the civilian, non-institutionalized US population. Information on demographics, serum measurements, lifestyle factors, sleep pattern (Question: How does it usually take to fall asleep at bedtime? More details via http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/SLQ_D.htm) and urinary environmental chemical concentrations was obtained by household interview. In the current analysis, the 2005–2006 cohort as the most recent study cohort with all the required information mentioned above was selected for statistical analysis (more details via http://wwwn.cdc.gov/nchs/nhanes/search/nhanes05_06.aspx). Informed consents were obtained from participating subjects by the NHANES researchers.

Biomonitoring

Urines were only collected in a representative subsample within 10 days of the household interview, being one third of the whole study cohort (more details via http://www.cdc.gov/nchs/data/nhanes/nhanes_05_06/lab_d.pdf), to measure environmental chemical concentrations in urine among people aged 6 and above (more details via http://www.cdc.gov/nchs/data/nhanes/nhanes_03_04/environmentalhealth_03.pdf). Urine specimens were processed, stored under appropriate frozen (−20 °C) conditions, and shipped to the Division of Environmental Health Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention for analysis.

The inductively coupled plasma-mass spectrometry (ICP-MS) method (Mulligan et al. 1990) was used to measure the following 12 elements in urine: beryllium (Be), cobalt (Co), molybdenum (Mo), cadmium (Cd), antimony (Sb), cesium (Cs), tungsten (W), tin (Sn), strontium (Sr), manganese (Mn) thallium (TI), lead (Pb), and uranium (U). Urine samples are diluted 1 + 9 with 2% (v/v), double-distilled, concentrated nitric acid containing both iridium (Ir) and rhodium (Rh) for multi-internal standardization (more details via http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/UHM_D.htm). The test principle utilized high-performance liquid chromatography-electrospray ionization-tandem mass spectrometry (HPLC-ESI-MS/MS) for the quantitative detection in urine of the following phthalate metabolites: monomethyl phthalate (MMP), monoethyl phthalate (MEP), monobutyl phthalate (MBP), monoisobutyl phthalate (MIBP), mono(3-carboxypropyl) phthalate (MCPP), mono(2-ethylhexyl) phthalate (MEHP), monobenzyl phthalate (MBzP), monoisononyl phthalate (MNP), mono(2-ethyl-5-oxohexyl) phthalate (MEOHP), mono(2-ethyl-5-hydroxyhexyl) phthalate (MEHHP), mono(2-ethyl-5-carboxypentyl) phthalate (MECPP), monocarboxyoctyl phthalate (MCOP), monocarboxynonyl phthalate (MCNP), and cyclohexane-1,2-dicarboxylic acid-mono(hydroxy-isononyl) ester (MHNCH). Urine samples are processed using enzymatic deconjugation of the glucuronidated metabolites followed by on-line solid phase extraction (SPE) coupled with reversed phase HPLC-ESI-MS/MS. Assay precision is improved by incorporating isotopically labeled internal standards of the phthalate metabolites and MHNCH (more details via http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/PHTHTE_D.htm).

Total and specific urine arsenic concentrations were determined by using inductively coupled-plasma dynamic reaction cell-mass spectrometry (ICP-DRC-MS) (more details via http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/UAS_D.htm) Urine is analyzed because urinary excretion is the major pathway for eliminating arsenic from the mammalian body. The method used on-line SPE coupled to HPLC and tandem mass spectrometry (HPL/CMS/MS). A sensitive method, SPE coupled on-line to HPLC and tandem mass spectrometry (MS/MS), was used for measuring bisphenol A (BPA), 4-tert-octylphenol (tOP), benzophenone-3 (BP-3), one chlorophenols triclosan, four parabens, and pesticides (more details via http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/EPH_D.htm and http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/PP_D.htm). A quantitative procedure was used for the measurement of nitrate, perchlorate, and thiocyanate in human urine using ion chromatography coupled with electrospray tandem mass spectrometry. Chromatographic separation is achieved using an IonPac AS16 column with sodium hydroxide as the eluant. The eluant from the column is ionized using an electrospray interface to generate and transmit negative ions into the mass spectrometer (more details via http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/PTH_D.htm).

To detect and measure metabolites of polyaromatic hydrocarbons (more details via http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/PAH_D.htm), the procedure involved enzymatic hydrolysis of glucuronidated/sulfated OH-polyaromatic hydrocarbons metabolites in urine, extraction, derivatization, and analysis using isotope dilution capillary gas chromatography tandem mass spectrometry (GC-MS/MS). Ion transitions specific to each analyte and carbon-13 labeled internal standards are monitored, and the abundances of each ion are measured. Moreover, solid phase extraction coupled to high-performance liquid chromatography-turbo ion spray ionization-tandem mass spectrometry (online SPE-HPLC-TIS-MS/MS) was used for the quantitative detection of perfluorooctane sulfonamide (PFOSA), 2-(N-methyl-perfluorooctane sulfonamido) acetic acid (Me-PFOSA-AcOH), 2-(N-ethyl-perfluorooctane sulfonamido) acetic acid (Et-PFOSA-AcOH), perfluorobutane sulfonate (PFBuS), perfluorohexane sulfonate (PFHxS), perfluorooctane sulfonate (PFOS), perfluoroheptanoate (PFHpA), perfluorooctanoate (PFOA), perfluorononanoate (PFNA), perfluorodecanoate (PFDeA), perfluoroundecanoate (PFUA), and perfluorododecanoate (PFDoA) (more details via http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/PFC_D.htm).

Statistical analysis

Americans aged 18–85 with all the required study variables were included in the current statistical analysis. Urinary environmental chemical concentrations were all log transformed because they were highly skewed to one side. Associations of urinary environmental chemical concentrations (x variables) and the five common sleep troubles in the past month (y variables; more details via http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/SLQ_D.htm) were examined by using survey-weighted logistic regression models, with P < 0.05 considered statistically significant. Covariates including urinary creatinine, age, sex, ratio of family income to poverty (proxy of socioeconomic status), body mass index, serum cotinine (biomarker of smoking status), vitamin D levels (more details via http://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/VID_D.htm), and physical activity level were adjusted. Statistical software STATA version 13.0 (STATA, College Station, Texas, USA; more details via http://www.stata.com/) was used to conduct all the analyses.

Ethics concerns

Since there are only secondary data analyses employed without any participant personal information identified by extracting statistical data from the NHANES website in the present study, no further ethics approval for conducting the present study is required (more details via http://www.ethicsguidebook.ac.uk/Secondary-analysis-106).

Results

Table 1 shows characteristics of the included study participants. Of all 5563 Americans aged 18–85, 2331 (42.0%) had wake-up at night, 2914 (52.5%) felt unrested during the day, 740 (13.4%) had leg jerks while sleeping, and 1059 (19.1%) had leg cramps while sleeping for 2+ times a month. Table 2 lists the frequency of common sleep troubles in the study participants. Tables 3, 4, 5, and 6 present the associations between urinary chemical concentrations and common sleep troubles, except for feeling overly sleeping during the day with no association found, accordingly. To be specific, higher levels of urinary arsenic, phthalates, and polyfluoroalkyl compounds were associated with wake-up at night. Higher levels of urinary 4-tert-octylphenol and polyfluoroalkyl compounds were associated with being unrested during the day. Higher levels of urinary arsenic, polyaromatic hydrocarbons, and polyfluoroalkyl compounds were associated with leg jerks while sleeping. Higher levels of urinary pesticides, heavy metals, phthalates, and polyaromatic hydrocarbons were associated with leg cramps while sleeping. There were no significant associations with other environmental chemicals such as parabens, bisphenol A, benzophenone-3, triclosan, perchlorate, nitrate, or thiocyanate (data not shown).

Table 1 Characteristics of the study participants aged 18–85 (N = 5563)
Full size table
Table 2 Characteristics of sleep troubles in the study participants aged 18–85 (N = 5563)
Full size table
Table 3 Associations between urinary chemicals and wake-up at night in adults
Full size table
Table 4 Associations between urinary chemicals and unrested during the day in adults
Full size table
Table 5 Associations between urinary chemicals and leg jerks while sleeping in adults
Full size table
Table 6 Associations between urinary chemicals and leg cramps while sleeping in adults
Full size table

Discussion

Arsenic and sleep

Low level of arsenic exposure was previously observed to be associated with sleep disturbance in copper smelter workers (n = 680; Lilis et al. 1985). Long-term poisoning was also found to affect sleep disorder in children (Ishi and Tamaoka 2015). In the present study, higher levels of urinary arsenic were found to be associated with wake-up at night and leg jerk while sleeping. However, to date no experimental research was available to confirm the biological mechanism yet.

Pesticides and sleep

Pesticide exposure, particularly 2,4,5-trichlorophenol, was recently found to influence idiopathic REM sleep behavior disorder in older adults (n = 694; Postuma et al. 2012; Neuberger et al. 1998). In the limited experimental research, exposure to pesticides seemed to reduce sleep time in male mice (21–24 g; Chaturvedi 1993). In the present study using national representative human sample, 2,5-dichlorophenol and 2,4-dichlorophenol showed borderline associations with leg cramps in sleeping. This would need future longitudinal or experimental research to confirm or refute the findings in order to rule out the statistical significance by chance.

Heavy metals and sleep

Uranium might directly affect the brain. Previously, it was observed that chronic uranium exposure (40 mg l(−1) in drinking water, for 90 days) increased in rapid eye movement sleep in rats (Lestaevel et al. 2005). In the present study, the consistent finding was shown in higher levels of urinary uranium and urinary antimony and leg cramps in sleeping.

Phthalates and sleep

The intravenous or intraperitoneal administration of bis(2-ethylhexyl) phthalate could prolong hexobarbital sleep time in mice and rats by enlarging the lipophilic pool (Swinyard et al. 1976). In the present study, monocyclohexyl showed an association with wake-up at night, and mono(carboxynonyl) and mono(3-carboxypropyl) showed associations with leg cramps in sleeping. These would also need future longitudinal or experimental research to confirm or refute the findings in order to rule out the statistical significance by chance.

Polyaromatic hydrocarbons and sleep

There was no direct research evidence on the relationship of polyaromatic hydrocarbons and sleep health. In the present study, it is the first time to show that higher urinary polyaromatic hydrocarbons, such as 2-hydroxyfluorene, 9-hydroxyfluorene, 2-hydroxyphenanthrene, and 1-hydroxypyrene, were significantly associated with leg cramps in sleeping. Similarly, urinary 2-hydroxyfluorene, 3-hydroxyfluorene, 1-hydroxypyrene were significantly associated with leg jerks in sleeping. Levels of other known urinary polyaromatic hydrocarbons were also higher in people with sleep disturbances, but the statistical significance was not reached.

Polyfluoroalkyl compounds and sleep

There were no indications that there could be any association between polyfluoroalkyl compounds and sleep health in the past. Polyfluoroalkyl compounds in human health is an under-studied area, mostly with a focus on fetal growth, neuro-behavioral problems in infants or children, or cognitive function in older adults (Bach et al. 2015; Donauer et al. 2015; Power et al. 2013; Maisonet et al. 2012; Hoffman et al. 2010). In the present study using national representative human sample, the new observations have consistently shown higher levels of urinary 2-(N-ethyl-PFOSA) acetate and perfluorododecanoic acid in people with wake-up at night, higher levels of urinary 2-(N-methyl-PFOSA) acetate, perfluorobutane sulfonic acid in people feeling unrested during the day, and higher levels of 2-(N-methyl-PFOSA) acetate, perfluorobutane sulfonic acid and perfluoroheptanoic acid in people with leg jerks in sleeping. However, to date there has been no experimental research to confirm yet.

Strengths and limitations

The present study has a few strengths. Firstly, this study was conducted in a large and nationally representative human sample with mixed ethnicities and socioeconomic status. Secondly, this was the first time to examine the risk associations between different sets of environmental chemical concentrations and common sleep troubles. However, there are still some limitations that cannot be ignored. First, there could be still other emerging chemicals from the living environments that we might not yet know and would need future research to further identify and examine. Second, objective measurements on sleep patterns were not available. Third, the causality cannot be established in the present study due to the cross-sectional study design in nature. Therefore, future studies with longitudinal or experimental study designs to confirm or refute the current findings and, if at all, to understand the persisting risk effects along the life course from those mentioned above environmental chemicals would be warranted.

Directions for future research, practice, and policy

Urinary arsenic, pesticides, heavy metals, phthalates, polyaromatic hydrocarbons and polyfluoroalkyl compounds were associated with common sleep troubles in adults For future research, studies understanding the biological mechanism and monitoring the risk effects along the life course with a longitudinal or experimental approach would be needed. For practice and policy, elimination of these environmental chemicals in order to improve sleep health might be considered.