Consideration of Age-Related Changes in Behavior Trends in Older Adults in Assessing Risks of Environmental Exposures

Abstract

Objectives: To explore age-related behavior differences between older and younger adults, and to review how older adult activity patterns are considered in evaluating the potential risk of exposure to environmental pollutants. Methods: Activity pattern data and their use in risk assessments were analyzed using the U.S. EPA Exposure Factors Handbook (EFH), U.S. EPA Consolidated Human Activity Pattern Database (CHAD), and peer-reviewed literature describing human health risk assessments. Results: The characterization by age of some factors likely to impact older adults’ exposures remains limited. We demonstrate that age-related behavior trends vary between younger and older adults, and these differences are rarely explicitly considered in environmental health risk assessment for older adults. Discussion: Incorporating older adult exposure factors into risk assessments may be challenging because of data gaps and difficulty in defining and appropriately binning older adults. Additional data related to older adult exposure factors are warranted for evaluating risk among this susceptible population.

Keywords

activity patterns behavior exposure factors older adults risk assessment

Introduction

In order to adequately protect public health, government agencies must assess the risks associated with exposures to myriad environmental chemicals and other stressors to determine which potential hazards are most significant. These human health risk assessments require information on both the hazard of a given pollutant, including the relationship between dose and effects, as well as an evaluation of potential exposures. Age-related changes in behavior and physiological function are important to consider in conducting human health risk assessments because these changes can affect both the exposure and physiological response to environmental contaminants. One source that presents these age-related changes in behaviors and some physiological factors (e.g., inhalation rates) is the U.S. Environmental Protection Agency’s (U.S. EPA) Exposure Factors Handbook (EFH). First published in 1989, the EFH provides summary statistics of human activity patterns and characteristics, or exposure factors, used to assess exposure and risk, and presents recommended values for each exposure factor (Moya & Phillips, 2002). Initially expanded and updated in 1997 (U.S. EPA, 1997), the EFH was recently revised and updated again, with the final EFH released in September of 2011 (U.S. EPA, 2011). Exposure factors in the 2011 EFH include drinking water ingestion, soil and dust ingestion, inhalation rate, skin surface area, various food intake values and activity factors (e.g., time spent outdoors and indoors). Users of these recommended values include various health, government and international organizations, such as the American Medical Association, U.S. EPA’s Office of Solid Waste, and Health Canada (Moya & Phillips, 2002; Phillips & Moya, 2013).

The recommended exposure factor values are presented by age-defined bins in the EFH. Based on the available data for each exposure factor, the number and age range of these bins vary for children and adults, with data on children generally divided into finer age bins than data presented for adults (U.S. EPA, 1997; U.S. EPA, 2011). The childhood age bins are based on the rapid behavioral and physiological changes and the impact these changes have on oral, dermal, and inhalation exposures (U.S. EPA, 2005a; U.S. EPA, 2008).

Likewise, older adults also undergo changes in behavior and physiology that may affect their exposure and health risk to environmental pollutants. Threats to mobility, such as declining physical, cognitive, and sensory functioning, become more prevalent with age (Stalvey, Owsley, Sloane, & Ball, 1999; U.S. EPA, 2007). As a result of limited mobility, older adults may spend prolonged periods of time in a single microenvironment. Although older adults may be less likely to encounter certain exposures than younger adults (e.g., traffic pollution), a single microenvironment may account for the majority of the exposure to environmental pollutants experienced by older adults (U.S. EPA, 2007). For example, older adults are likely to spend more time indoors, where the concentration of hazardous air pollutants may be greater (Geller & Zenick, 2005). Physiologically, increases in heart disease, stroke, hypertension, diabetes, and neurologic diseases, along with decreases in weight, lean muscle mass, energy utilization, organ function, hepatic and renal clearance, antioxidant and glutathione levels, and cellular defense mechanisms, become more prevalent with age (Boss & Seegmiller, 1981; Pleis, Lucas, & Ward, 2009; Schoenborn & Heyman, 2009; U.S EPA, 2007). Declines in respiratory and cardiovascular function have also been observed in older adults even without disease (U.S EPA, 2005b). Changes in the various organ systems may impact the absorption, distribution, metabolism, and excretion of chemicals in the body (Geller & Zenick, 2005; U.S. EPA, 2005b). Therefore, older adults may experience differential exposures and risks to evironmental contaminants. With the older adult population increasing, from a public health perspective, there is a growing need to understand how this population is affected by environmental exposures, what potential health risks exist, and how to mitigate these risks. The purpose of this article is to explore how exposure factor trends, with a particular focus on behavioral factors, differ between younger and older adults and how older adult behavior is typically considered in risk assessments. Although this article will analyze several specific behavioral factors, it does not present a comprehensive evaluation of differences in exposure factors between younger and older adults.

Method

Several different sources of information were utilized in reviewing age-related changes in exposure factors among older adults and the use of these data in conducting exposure and risk assessments. As described below, these primarily include (1) the 1997 EFH and 2011 EFH, (2) the U.S. EPA Consolidated Human Activity Pattern Database (CHAD), (3) the Summary Report of a Peer Involvement Workshop on the Development of an Exposure Factors Handbook for the Aging, and (4) peer-reviewed literature describing results of site-specific risk assessments. Since these sources of information define older adults differently, an overall cutoff age for older adults has not been explicitly defined for the purposes of this article.

1997 and 2011 Editions of the Exposure Factors Handbook

The 1997 EFH and 2011 EFH were reviewed to identify changes over time to the definition of age bins for older adults, with comparisons made between the age binning for older adults and children (see Table 1). The EFHs were also used to develop search terms and search strategies for conducting literature searches related to exposure factors used in site-specific risk assessments.

Table 1.

Comparison of Children and Adult Age Bins for a Majority of the Exposure Factors in the Exposure Factors Handbook 2011 Update (U.S. EPA, 2011).

Exposure factor	Children	Adults
Drinking water ingestion	Birth to <1 month	≥21 yrs
	1 to <3 months	>65 yrs
	3 to <6 months
	6 to <12 months
	1 to <2 yrs
	2 to <3 yrs
	3 to <6 yrs
	6 to <11 yrs
	11 to <16 yrs
	16 to <18 yrs
	18 to <21 yrs
Soil and dust ingestion	6 weeks to <12 months	Adult
	1 to <6 yrs
	3 to <6 yrs
	6 to <21 yrs
Exposure inhalation	Birth to <1 month	21 to <31 yrs
	1 to <3 months	31 to <41 yrs
	3 to <6 months	41 to <51 yrs
	6 to <12 months	51 to <61 yrs
	Birth to <1 yr	61 to <71 yrs
	1 to <2 yrs	71 to <81 yrs
	2 to <3 yrs	≥81 yrs
	3 to <6 yrs
	6 to <11 yrs
	11 to <16 yrs
	16 to <21 yrs
Body surface area	Birth to <1 month	Total Body:
	1 to <3 months	21-30 yrs
	3 to <6 months	30 to <40 yrs
	6 to <12 months	40 to <50 yrs
	1 to <2 yrs	50 to <60 yrs
	2 to <3 yrs	60 to <70 yrs
	3 to <6 yrs	70 to <80 yrs
	6 to <11 yrs	≥80 yrs
	11 to <16 yrs
	16 to <21 yrs
Body weight	Birth to <1 month	Body Parts:
	1 to <3 months	≥21 yrs
	3 to <6 months	21 to <30 yrs
	6 to <12 months	30 to <40 yrs
	1 to <2 yrs	40 to <50 yrs
	2 to <3 yrs	50 to <60 yrs
	3 to <6 yrs	60 to <70 yrs
	6 to <11 yrs	70 to <80 yrs
	11 to <16 yrs	≥80 yrs
	16 to <21 yrs	≥80 yrs
Total fruit, vegetable, meat, dairy, and grain intake	Birth to <1 yr	20 to <50 yrs
	1 to <2 yrs	≥50 yrs
	2 to <3 yrs
	3 to <6 yrs
	6 to <11 yrs
	11 to <16 yrs
	16 to <21 yrs
Individual fruits, vegetables, meats, dairy, and grains intake	Birth-1 yr	20-49 yrs
	1-2 yrs	Females 13-49 yrs
	3-5 yrs	≥50 yrs
	6-12 yrs
	13-19 yrs
Fish intake	Birth-1 yr	20-49 yrs
	1 to <2 yrs	Females 13-49 yrs
	2 to <3 yrs	≥50 yrs
	3 to <6 yrs
	6 to <11 yrs
	11 to <16 yrs
	16 to <21 yrs
Total fats intake	Birth to <1 month	21 to <31 yrs
	1 to <3 months	31 to <41 yrs
	3 to <6 months	41 to <51 yrs
	6 to <12 months	51 to <61 yrs
	1 to <2 yrs	61 to <71 yrs
	2 to <3 yrs	71 to <81 yrs
	3 to <6 yrs	≥81 yrs
	6 to <11 yrs
	11 to <16 yrs
	16 to <21 yrs
Home-produced food intake	1-2 yrs	20-39 yrs
	3-5 yrs	40-69 yrs
	6-11 yrs	≥70 yrs
	12-19 yrs	≥70 yrs
Total food intake	Birth to <1 month	20 to <40 yrs
	1 to <3 months	40 to <70 yrs
	3 to <6 months	≥70 yrs
	6 to <12 months
	1 to <2 yrs
	2 to <3 yrs
	3 to <6 yrs
	6 to <11 yrs
	11 to <16 yrs
	16 to <21 yrs
Activity factors: Time spent indoors and outdoors	Birth to <1 month	18 to <65 yrs
	1 to <3 months	≥65 yrs
	3 to <6 months
	6 to <12 months
	1 to <2 yrs
	2 to <3 yrs
	3 to <6 yrs
	6 to <11 yrs
	11 to <16 yrs
	16 to <21 yrs

Identifying Activity Pattern Data

Activity pattern data from CHAD were examined to evaluate differences in age-related trends in activity patterns between younger and older adults and to examine variability in activity patterns among older adults. Since only selected activities from CHAD were evaluated, this article does not provide a comprehensive evaluation of activity pattern data.

CHAD includes personal activity information collected from 12 human activity surveys in the United States and contains 875,338 records of personal activity diaries on an hourly basis, representing 144 activities at 115 locations (McCurdy, Glen, Smith, & Lakkadi, 2000). CHAD has been utilized in conducting quantitative exposure assessments as a part of the reviews of the National Ambient Air Quality Standards, in exposure models such as Stochastic Human Exposure and Dose Simulation for Particulate Matter (SHEDS-PM), and in the derivation of age-specific and activity-specific inhalation rates in the EFH (Burke, Zufall, & Ozkaynak, 2001; U.S. EPA, 2011). In our work, personal activity diaries for participants 18 and older were extracted from CHAD for further analyses. In total, 10,961 participants were included in our analysis, with a total of 482,552 hourly activity records, representing 127 activities at 95 locations. Analyses of 21 activities and 15 locations are presented herein, which were selected based on the number of activity or location records available for older (61-99 years) adults (e.g., “Attend full-time school” and “At dry cleaners” were excluded due to small sample size), the relevancy of the activity or location to older adults (e.g., “Attend K-12” and “Work for fraternity organizations” were excluded due to irrelevancy), and the consistency of the number of records available across the older adult age group for each activity or location. For instance, if many activity records were available for the youngest and oldest older adult ages, and records for the intermediate older adult ages were limited, then the activity was excluded from further analysis (e.g., “Think and relax” and “Travel for active leisure” were excluded due to inconsistency). A comprehensive list of activities and locations presented in CHAD that were excluded from the analysis is presented in Appendix A.

Percent of daily time a person spent on an activity or at a location was calculated for each participant. Summary statistics of percent of daily time spent engaged in specific activities or at specific locations were used to examine the differences in activity patterns between younger (18-60 years) and older (61-99 years) adults. Respective associations between age and activity patterns were examined for the younger and older adult age groups with Spearman correlation coefficients to identify age-related trends in activity patterns, that is, whether people tend to spend more or less time on an activity as they age. Differences in Spearman correlation coefficients in both direction (positive and negative), and statistical significance, were also examined to determine whether age-related trends differ between younger and older adults. Although the cutoff age of 60 years roughly corresponds to the employed population and is consistent with the age bins presented for some exposure factors in the 2011 EFH (Table 1), we acknowledge that in differentiating between younger and older adults, this cutoff age is somewhat arbitrary. Therefore, activity pattern variation with age as a continuous function was also examined with a LOESS (Local Polynomial Regression Fitting) analysis. This nonparametric regression method provides great flexibility and does not assume a parametric form for the regression surface. The smoothing parameter was determined based on the optimization of the bias-corrected Akaike information criterion, and the robustness of the nonparametric fit was improved by iterative reweighting. The LOESS analysis was performed for a total of six activities, where there was either a difference or no difference between younger and older adults in both the direction and statistical significance of the correlation coefficients. Statistical analyses were conducted with the Statistical Analysis System (SAS version 9.1, SAS Inc, Cary, NC). The significance level was 0.05 except where specified.

Peer Involvement Workshop on the Development of an Exposure Factors Handbook for the Aging

On February 14 to 15, 2007, the U.S. EPA’s National Center for Environmental Assessment hosted a workshop in Arlington, VA, which focused on addressing charge questions concerning the availability of activity pattern and behavior information on the aging, data gaps and research needs, the appropriateness of age bins for the aging when conducting exposure/risk assessments, potential users of an EFH for the aging, and the most beneficial presentation of data for exposure/risk assessors (U.S. EPA, 2007). The workshop’s 30 attendees (20 invited expert panelists and 10 observers) consisted of individuals from the U.S. EPA, other federal agencies, state governments, academia, and industry. Following completion of the workshop, a workshop summary report was developed that included information on the identification of data gaps, research needs, and recommendations for how exposure factors specific to older adults should be considered in conducting exposure and risk assessments (U.S. EPA, 2007). This summary report was reviewed to identify sources of older adult behavior data, data gaps, and challenges in defining and appropriately binning older adults.

Site-Specific Risk Assessments

A literature search was conducted to identify site-specific risk assessments from the early 1990s to 2009 that used exposure factors similar to those presented in the EFHs. These risk assessments were reviewed to determine which populations are typically examined, which populations are considered susceptible, the frequency with which older adults are examined, and whether older adults are explicitly considered as a susceptible population. Risk assessment literature was identified through multiple searches in PubMed and ScienceDirect using combinations of the following terms: Risk assessment, site specific risk assessment, community based risk assessment, human health, site contamination, water contamination, soil contamination, food contamination, and Superfund.

Results

Presentation of Exposure Factors by Age Bins in the EFHs

Several changes have been made to the EFH since the publication of the 1997 EFH. The 2011 EFH incorporates new exposure factors, published literature since 1997, and data from a Child-Specific EFH finalized by EPA in 2008 (U.S. EPA, 2008). Although many improvements have been made that reflect the current increased knowledge about human variability in behavior and physiology, the binning of exposure factors for older adults remains limited for some exposure factors in the 2011 EFH. For example, the 1997 EFH presents one recommended drinking water ingestion value for adults, defined as ≥ 20 years old. In the 2011 EFH, adults are divided into two age bins, ≥ 21 and ≥ 65 years old. Multiple age bins for older adults have been added to several exposure factors in the 2011 EFH (e.g., body surface area, inhalation rates, fish intake rates, and body weight). As an example, for long-term inhalation rates, the 2011 EFH presents older adult information by bins of 51 to < 61, 61 to < 71, 71 to < 81 and ≥ 81 years old. In the 1997 EFH, there is no age bin for older adults, but instead a single bin, ≥ 19 years old, for all adults (U.S. EPA, 1997; U.S. EPA, 2011).

Table 1 presents the children and adult age bins for a majority of the exposure factors in the 2011 EFH. Although there is an overall increase in age-defined information for children and adults in the 2011 EFH, children are divided into more finely defined age bins than older adults. The availability of data for children and the “Guidance on Selecting Age Groups for Monitoring and Assessing Childhood Exposures to Environmental Contaminants” report can account for these finely defined age bins (U.S. EPA, 2005a; U.S. EPA, 2008). Data in the 2011 EFH comes from a variety of sources, mostly from the analyses of the individual study authors (U.S. EPA, 2011). Therefore, the lack of older adult information is mainly due to the availability of data for more finely defined adult age groups in the published literature and lack of guidance on age binning for adults. In addition, variability in the population for some exposure factors may be more influenced by parameters other than age. For example, incidental ingestion of soil and dust by adults may be more dependent on activities and microenvironments than by the age of the individual.

Activity Pattern Data: CHAD

Age-related trends in activity patterns were examined by calculating Spearman correlation coefficients between age and the mean percent daily time spent on 21 activities and at 15 locations for both younger and older adults (Tables 2 -5). Between younger and older adults, there is a difference in the direction of the correlation coefficients for 13 activities and 8 locations (see Tables 2 and 4), indicating a reversal of the age-related trend in time spent on an activity or at a location. The directional difference of correlation coefficients is important because these differences could indicate potential exposure differences between younger and older adults. For instance, for time spent at “your residence-other outdoor,” the correlation coefficients for younger and older adults are 0.11 (p<0.0001) and −0.09 (p = 0.025), respectively (see Table 4). The time spent outdoors may be indicative of the time spent on outdoor activities, such as gardening and participating in sports. Therefore, younger and older adults may differ in whether they progressively increase or decrease time spent on outdoor activities as they age, affecting their inhalation of road pollution, soil ingestion, soil dermal contact, and other outdoor-related exposures.

Table 2.

Spearman Correlation Coefficients (r_S) and Statistical Significance (p-value) of the Association Between Age and Daily Time Spent on an Activity for Younger and Older Adults.

	Younger adults			Older adults
Activity	N	r_S	p-value	N	r_S	p-value
Attend a party	223	0.06	0.340	50	−0.15	0.285
Attend sports events	163	−0.02	0.803	14	0.08	0.776
Care for pets/animals	635	0.03	0.390	219	0.06	0.344
Care of clothes	1120	0.06	0.060	277	−0.03	0.649
Care of plants	189	0.14	0.056	144	−0.09	0.262
Clean house	1458	0.07	0.009	646	−0.01	0.883
Clean-up food	1058	0.01	0.728	462	−0.06	0.221
Eat	7567	0.13	<.0001	2223	0.02	0.274
Exercise	397	−0.26	<.0001	95	−0.21	0.043
Indoor chores	771	0.08	0.026	148	−0.01	0.936
Outdoor chores	761	0.10	0.008	300	0.00	0.964
Participate in sports	810	−0.16	<.0001	238	−0.24	0.000
Practice religion	552	−0.11	0.012	312	0.01	0.802
Prepare food	4114	0.07	<.0001	1309	−0.05	0.053
Read newspaper	877	0.07	0.036	645	0.09	0.026
Shop for clothes or household goods	1252	−0.01	0.748	348	−0.12	0.022
Shop for food	1151	0.12	<.0001	362	0.07	0.210
Shower, bathe	3870	−0.03	0.098	858	0.07	0.038
Sleep or nap	8439	−0.11	<.0001	2310	0.14	<.0001
Watch television	5132	0.04	0.006	1815	0.14	<.0001
Work, income-related only	4451	0.04	0.004	367	−0.26	<.0001

Table 3.

Differences in the Direction and Statistical Significance of Spearman Correlation Coefficients Between Younger and Older Adults in the Association Between Age and Daily Time Spent on an Activity.

	Difference in significance?
		Yes	No
Difference in direction?	Yes	Clean house	Attend a party
		Indoor chores^a	Attend sports event
		Outdoor chores	Care of clothes
		Practice religion	Care of plants
		Prepare food^a	Clean-up food
		Shower, bathe^a*	Sleep or nap*
		Work, income-related only*	Sleep or nap*
	No	Eat	Care for pets/animals
		Shop for clothes or household goods*	Exercise^a*
		Shop for clothes or household goods*	Participate in sports*
		Shop for food	Read newspaper*
		Shop for food	Watch TV^a*

Selected for the LOESS (Local Polynomial Regression Fitting) analysis

Correlation coefficient is statistically significant (p≤0.05) for older adults

Table 4.

Spearman Correlation Coefficients (r_S) and Statistical Significance (p-value) of the Association Between Age and Daily Time Spent at A Location for Younger and Older Adults.

	Younger adults			Older adults
Location	N	r_S	p-value	N	r_S	p-value
At restaurant	2389	0.01	0.801	504	0.08	0.085
Basement	158	−0.15	0.060	62	0.01	0.921
Bathroom	5107	−0.02	0.142	1255	0.08	0.005
Bedroom	6632	−0.11	<.0001	2045	0.12	<.0001
Dining room	1381	0.10	0.000	510	0.07	0.127
Grocery store/convenience store	1890	0.03	0.213	486	0.04	0.350
Hospital/health care facility/doctor’s office	738	−0.12	0.001	235	−0.05	0.446
Kitchen	5078	0.16	<.0001	1825	0.04	0.079
Living room/family room	5439	0.05	<.0001	1855	0.22	<.0001
Public building/library/museum/theater	566	−0.01	0.813	140	0.07	0.406
Residence, outdoor	262	0.07	0.264	52	−0.03	0.808
Travel by car	6431	−0.01	0.575	1515	−0.04	0.163
Travel by walk	1461	−0.02	0.403	396	0.05	0.331
Within 10 yards of street	732	−0.03	0.369	117	0.11	0.253
Your residence—other outdoor	1500	0.11	<.0001	653	−0.09	0.025

Table 5.

Differences in the Direction and Statistical Significance of Spearman Correlation Coefficients Between Younger and Older Adults in the Association Between Age and Daily Time Spent at a Location.

	Difference in significance?
		Yes	No
Difference in direction?	Yes	Bathroom*	Basement
			Bedroom*
			Public building/library/museum/theatre
			Residence, outdoor
			Travel by walk
			Within 10 yards of street
			Your residence—other outdoor*
	No	Dining room	At restaurant
		Hospital/health care facility/doctor’s office	Grocery store/convenience store
		Kitchen	Living room/family room*
		Kitchen	Travel by car^a

Selected for the LOESS (Local Polynomial Regression Fitting) analysis

Correlation coefficient is statistically significant (p≤0.05) for older adults

Correlation coefficients were also calculated by gender, race, housing, employment, fulltime employment, and asthma prevalence. The coefficients were then stratified by the younger and older adult age groups. These results were not presented because the sample size for each covariate was small after stratification and lacked the power to accurately examine the associations between age and activity patterns. However, the stratification of the activity and location data by covariates is important for examining behavior across and within various groups.

Six activities and locations, which either had a difference or no difference in both the direction and statistical significance of the correlation coefficients between younger and older adults (see Tables 3 and 5), were selected for the LOESS analysis. Activities and locations included in the LOESS analysis reflect two extreme scenarios in Tables 3 & 5, (1) there are both directional and statistical significance differences in the Spearman correlation coefficients between older and younger adults; and (2) there are neither directional nor statistical significance differences in the Spearman correlation coefficients between older and younger adults. The two scenarios are illustrated with six activities or locations presented in Figure 1, and the rest of the activities and locations falling into the two scenarios were also analyzed with LOESS and presented in Appendix B. The LOESS analysis further demonstrates the activity variation between younger and older adults and among older adults (Figure 1). For some activities, there is a continuous positive or negative association between age and time spent on an activity across all ages. For other activities, the association between age and time spent on the activity varies based on which age range is examined. It is important to note, however, that the behavior of a small sample size may not describe the average behavior of the represented population. Therefore, limited data on older adults in multiple comparisons may produce an inaccurate description of older adult behavior and may not be able to fully capture behavior variability among older adults.

Figure 1.

Association between age and the mean percent of daily time spent on six activities from CHAD by LOESS (Local Polynomial Regression Fitting) analysis.

Site-Specific Risk Assessments

The extent to which older adults and their age-related behavior changes are considered in human health risk assessments was analyzed through examining publications describing 60 different site-specific risk assessments. These risk assessments used recommended exposure factor values from the EFH and other exposure factor related data to determine contaminant exposures. Based on these risk assessments, the need for additional and/or more useful older adult exposure related data available to risk assessors was evaluated.

Out of the 60 risk assessments reviewed, only one study, Diaz and Dominguez (2009), specifically examined the risk to older adults. This study examined the risk to 2- to 6- and 9- to 12-year-old children and 70-year-old adults, or the “most sensitive groups,” from inhaling particulate matter less than 2.5 micrometers in diameter (i.e., PM_2.5) in Mexico City. Exposure to these particles has been observed to be positively associated with respiratory and cardiovascular morbidity and mortality. Other studies either examined broadly defined adult age bins (e.g., 26-82, 15-70 and ≥ 18 years old), or concluded that older adults are potentially more susceptible, but did not present exposure factor values or risk results specifically for older adults in their assessments (Asmus et al., 2008; Hang et al., 2009; Nouwen et al., 2001; Turczynowicz, Fitzgerald, Nitschke, Mangas, & McLean, 2007; Urban, Tachovsky, Haws, Staskal, & Harris, 2009). Children, fishermen’s children and adult fishermen were the only populations where specific reasons for their examination were presented (Hacon, Rochedo, Campos, & Lacerda, 1997; Parsons, Huntley, Ebert, Algeo, & Keenan, 1991; Priha, Hellman, & Sorvari, 2005; Veiga, Amaral, & Fernandes, 1998;). Although older adults are widely recognized to constitute a potential at risk population, the paucity of data related to age-specific exposure factors among the aging, as well as limitations in defining older adults has clearly contributed to difficulties in evaluating risks to aging populations in these risk assessments.

Discussion and Conclusions

Based on our analysis of available site-specific risk assessments, it is clear that the extent to which older adults and their age-related behavior changes are considered in risk assessments is limited. The incorporation of older adult data into risk assessments may be challenging because of existing data gaps and difficulty in defining and appropriately binning older adults. In addition, the lack of available exposure-response information for older adults limits the implementation of the current knowledge regarding differential exposures experienced by older adults. The 2007 EFH Workshop identified information on water exposures, soil and dust exposure, product use around the home, hobbies, time spent in microenvironments, and time spent commuting as areas of limited data for older adults (U.S. EPA, 2007). The fulfillment of these data gaps would help further identify behavioral differences between younger and older adults and among older adults, and determine appropriate age bins.

Data gaps may also exist due to limitations in obtaining behavior data on older adults. A behavior diary, questionnaire, or interview may not be completed because of problems correlated with chronological age, such as cognitive or sensory impairments, handwriting or energy limitations, or poor health. Therefore, collected data may only represent those older adults who are the most healthy, cognitively intact, or capable of being interviewed or keeping track of their behavior (Lawton, 1999). Older adults not represented by this collected data may be at greatest risk to environmental contaminants due to their physiological or behavioral differences. It should be noted that innovative strategies, such as ecological momentary assessment, global positioning system (GPS), and direct performance measures, were developed in the past few years to collect personal activity patterns (Baird, Solcz, Gale-Ross, Blake, 2009; Moskowitz & Young, 2006; Oliver, Badland, Mavoa, Duncan, & Duncan, 2010). Personal activity patterns can be collected with improved accuracy (e.g., the ecological momentary assessment, and GPS) and less burden on the study participants (e.g., GPS) with these novel approaches. If widely adapted, these methods could potentially improve the collection and the availability of personal activity patterns for older adults.

Even if exposure-related data gaps on older adults were reduced, such data may not necessarily be readily useful to risk assessors. Chronological age may be the easiest and most convenient binning method, but binning older adults by chronological age alone presents several limitations. First, older adults are a heterogeneous group who age biologically at different rates. For some people, old age may start earlier than age 65, whereas for healthy adults, old age may not start until after age 65. Secondly, behavior is based more on biological than chronological age and is dependent on mobility, health status, employment/retirement, and so forth. Therefore, binning older adults by chronological age alone may be inappropriate (U.S. EPA, 2007).

On-going efforts by the U.S. EPA have been directed at researching ways to gather and present data for older adults. Literature searches have been conducted to investigate the availability of exposure data by functional status, lean muscle mass, and living situation. Functional status is measured using activities of daily living (ADL) or instrumental activities of daily living (IADL). These have been used to describe and compare the level of disability of older adults (Sonn, 1996). ADL is a term used to define the basic tasks of everyday life including eating, bathing, dressing, toileting, and transferring. IADLs refer to more complex activities than those included in the ADLs, such as handling personal finances, meal preparation, shopping, traveling, doing housework, using the telephone, and taking medications (Wiener, Hanley, Clark, & Van Nostrand, 1990). Ability to perform these activities can determine independence and therefore personal mobility, which in turn affects the exposures experienced by the individuals (Anttila, 1991). Studies in the literature also show correlations between muscle strength, walking speed, and physical activity (Aniansson, Rundgren, & Sperling, 1980). Weight gain during aging is also associated with decrease in function and mobility (Bannerman et al., 2002; Davison, Ford, Cogswell, & Dietz, 2002; Di Francesco et al., 2005; Guallar-Castillon et al., 2007; Launer, Harris, Rumpel, & Madans, 1994). Research shows that for older individuals, exposure levels may be affected by their ability to perform daily activities. Collection of exposure data by different microenvironments, which may be defined by ADL and IADL, may be a more appropriate way for characterizing activities and behaviors of older adults. More research in this area is necessary to fully understand the relationships among all the factors affecting exposure levels in older adults.

In spite of the efforts to collect activity pattern data, most existing data of personal activity patterns for older adults have not been collected in a systematic way, that is, exposure factors might not be collected altogether with other environmental or physiological factors. As a result, personal activities can only be classified into limited subgroups, resulting in imprecise exposure estimates for older adults. Some recent and ongoing longitudinal studies of aging illustrate the protocols to collect age-related data on activity patterns and environmental exposures (Chen & Wilmoth, 2004; Hauser & Weir, 2010; Kasper, Freedman, & Kalton, 2009). For example, during the Health and Retirement Study, personal activities were collected along with other parameters, such as health condition, social economic status (SES), and genome data (Chen & Wilmoth, 2004). Similarly, personal activities were collected along with detailed environmental and SES conditions during the ongoing National Health and Aging Trends Study (Kasper et al., 2009). Simultaneous collection of personal activity patterns and environmental or physiological parameters will enhance the estimate of personal activity patterns for different susceptible populations, and therefore improve personal exposure estimates for older adults.

The CHAD analysis demonstrates similarities and differences in age-related behavior trends between younger and older adults, and behavioral differences among older adults. For example, while a decrease in exercise with age was observed for adults of all ages (Figure 1b), age-related behavior trends differed between younger and older adults for time spent preparing food (Figure 1e) and time spent performing indoor and outdoor chores (Table 3). This is consistent with several publications utilizing data from the 1999 to 2008 National Health Interview Surveys (NHIS) that examined the relationship between age and leisure-time physical activity. The percentage of the survey population, consisting of adults either 18 to 65 years old or 18 to ≥ 75 years old, participating regularly in leisure-time physical activity decreased with age in each publication (Adams & Schoenborn, 2006; Barnes, 2007; Pleis et al., 2009; Pleis & Lucas, 2007; Schoenborn, Adams, Barnes, Vickerie, & Schiller, 2004). However, as demonstrated in the CHAD analysis, the relationship between age and other activities may not be as direct as the relationship between age and leisure-time physical activity. Ribas-Barba et al. (2007) observed significant age-related trends in food consumption among a random sample of individuals in Northeastern Spain (Evaluation of Nutritional Status in Catalonia (ENCAT) 1992-93 and ENCAT 2002-03), but also demonstrated that these trends may change over time. Therefore, the most current information on behavior may be necessary to most accurately determine potential exposures. In addition, it is important to consider other factors such as gender, mobility, health status, family and societal role, child responsibility, and retirement/employment that may influence activity patterns at an older age (Singleton, 1999; Stalvey et al., 1999; Vadarevu & Stopher, 1996).

The differences and similarities in age-related trends would not have been detected if these data were presented for a lumped group of all older adults or adults of all ages. The 2007 EFH Workshop concluded that older adults warrant special attention in an EFH, and recommended that older adult information be added to a future EFH so that exposure factor comparisons can be made between older adults and other populations (U.S. EPA, 2007). Based on previous and current uses of EFH information in risk assessments, additional data collection efforts focused on older adults would be beneficial to risk assessors. Consolidation of these data in future additions of the EFH can serve as a tool for understanding differential exposure among older adults. Consolidation of these data can also facilitate cumulative risk assessment for older adults, because the detailed exposure factors could be used to evaluate personal exposures to multiple environmental stressors and could be used to refine the estimation of personal exposure history (Sexton, 2012). The older adult population is quickly increasing in size and has been considered the most rapidly changing population. Therefore, the need to understand how this population is affected by environmental exposures and what potential health risks exist is becoming increasingly important to public health officials.

Footnotes

Appendix A. Activities and Locations Presented in CHAD That Were Excluded From the Analysis

Appendix B. LOESS Figures

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported in part by an appointment to the Research Participation Program for the U.S. Environmental Protection Agency, Office of Research and Development, administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and the EPA. The views expressed are those of the authors and do not necessarily reflect the views or policies of the U.S. EPA.

References

Adams

P. F.

Schoenborn

C. A.

(2006). Health behaviors of adults: United States, 2002-2004. National Center for Health Statistics. Vital and health statistics, 10(230), 1-140.

Aniansson

Rundgren

Sperling

(1980). Evaluation of functional capacity in activities of daily living in 70-year-old men and women. Scandinavian Journal of Rehabilitation Medicine, 12(4), 145-154.

Anttila

(1991). Functional capacity in two elderly populations aged 75 or over: Comparisons at 10-years’ interval. Journal of Clinical Epidemiology, 44, 1181-1186.

Asmus

Alonzo

Palácios

da Silva

A. P.

Filhote

Buosi

Câmara

(2008). Assessment of human health risk from organochlorine pesticide residues in Cidade don Meninos, Duque de Caxias, Rio de Janeiro, Brazil. Cadernos de Saúde Pública, 24(4), 755-766.

Baird

A. D.

Solcz

S. L.

Gale-Ross

Blake

T. M.

(2009). Older adults and capacity-related assessment: Promise and caution. Experimental Aging Research, 35, 297-316.

Bannerman

Miller

M. D.

Daniels

L. A.

Cobiac

Giles

L. C.

Whitehead

. . . Crotty

(2002). Anthropometric indices predict physical function and mobility in older Australians: The Australian longitudinal study of aging. Public Health Nutrition, 5, 655-662.

Barnes

(2007). Physical activity among adults: United States, 2000 and 2005 [Table 3, Age-adjusted percentages of leisure-time physical activity among adults 18 years and over, by selected characteristics: United States, 2000 and 2005].Hyattsville, MD: National Center for Health Statistics. Retrieved from http://www.cdc.gov/nchs/data/hestat/physicalactivity/physicalactivity.htm

Boss

G. R.

Seegmiller

J. E.

(1981). Age-related physiological changes and their clinical significance. Western Journal of Medicine, 135, 434-440.

Burke

J. M.

Zufall

M. J.

Ozkaynak

H. A.

(2001). Population exposure model for particulate matter: Case study results for PM_2.5 in Philadelphia, PA. Journal of Exposure Analysis and Environmental Epidemiology, 11, 470-489.

10.

Chen

P. C.

Wilmoth

J. M.

(2004). The effects of residential mobility on ADL and IADL limitations among the very old living in the community. Journals of Gerontology, 59B(3), S164-S172.

11.

Davison

K. K.

Ford

E. S.

Cogswell

M. E.

Dietz

W. H.

(2002). Percentage of body fat and body mass index are associated with mobility limitations in people aged 70 and older from NHANES III. Journal of the American Geriatrics Society, 50, 1802-1809.

12.

Diaz

R. V.

Dominguez

E. R.

(2009). Health risk by inhalation of PM_2.5 in the metropolitan zone of the City of Mexico. Ecotoxicology and Environmental Safety, 72, 866-871.

13.

Di Francesco

Zamboni

Zoico

Bortolani

Maggi

Bissoli

Bosello

(2005). Relationships between leisure-time physical activity, obesity and disability in elderly men. Aging Clinical and Experimental Research, 17(3), 201-206.

14.

Geller

A. M.

Zenick

(2005). Aging and the environment: A research framework. Environmental Health Perspectives, 113, 1257-1262.

15.

Guallar-Castillon

Sagardui-Villamor

Banegas

J. R.

Graciani

Fornes

N. S.

Garcia

E. L.

Rodriguez-Artalejo

(2007). Waist circumference as a predictor of disability among older adults. Obesity, 15(1), 233-244.

16.

Hacon

Rochedo

Campos

Lacerda

L. D.

(1997). Mercury exposure through fish consumption in the urban area of Alta Floresta in the Amazon Basin. Journal of Geochemical Exploration, 58(2-3), 209-216.

17.

Hang

Wang

Zhou

Chen

(2009). Risk assessment of potentially toxic element pollution in soils and rice (Oryza sativa) in a typical area of the Yangtze River Delta. Environmental Pollution, 157, 2542-2549.

18.

Hauser

R. M.

Weir

(2010). Recent developments in longitudinal studies of aging in the United States. Demography, 47(Suppl), S111-S130.

19.

Kasper

Freedman

Kalton

(2009). Overview of the National health and Aging Trends Study: Design, data, and dissemination. Gerontologist, 49, 520-520.

20.

Launer

L. J.

Harris

Rumpel

Madans

(1994). Body mass index, weight change, and risk of mobility disability in middle-aged and older women. The epidemiologic follow-up study of NHANES I. Journal of American Medical Association, 271, 1093-1098.

21.

Lawton

M. P.

(1999). Methods and concepts for time-budget research on elders. In Pentland

W. E.

Harvey

A. S.

Lawton

M. P.

McColl

M. A.

(Eds.), Time use research in the social sciences (pp. 107-125). New York: Kluwer Academic.

22.

McCurdy

Glen

Smith

Lakkadi

(2000). The national exposure research laboratory’s consolidated human activity database. Journal of Exposure Analysis and Environmental Epidemiology, 10(6 Part 1), 566-578.

23.

Moskowitz

D. S.

Young

S. N.

(2006). Ecological momentary assessment: What it is and why it is a method of the future in clinical phychopharmacology. Journal of Psychiatry and Neuroscience, 31(1), 13-20.

24.

Moya

Phillips

(2002). Overview of the use of the U.S. EPA exposure factors handbook. International Journal of Hygiene and Environmental Health, 205(1-2), 155-159.

25.

Nouwen

Cornelis

De Fré

Wevers

Viaene

Mensink

Guezens

(2001). Health risk assessment of dioxin emissions from municipal waste incinerators: The Neerlandquarter (Wilrijk, Belgium). Chemosphere, 43, 909-923.

26.

Oliver

Badland

Mavoa

Duncan

M. J.

Duncan

(2010). Combining, GPS, GIS, and Accelerometry: Methodological issues in the assessment of location and intensity of travel behaviors. Journal of Physical Activity and Health, 7(1), 102-108.

27.

Parsons

A. H.

Huntley

S. L.

Ebert

E. S.

Algeo

E. R.

Keenan

R. E.

(1991). Risk assessment for dioxin in Columbia river fish. Chemosphere, 23, 1709-1717.

28.

Phillips

Moya

(2013). The evolution of EPA’s Exposure factors handbook and its future as an exposure assessment resource. Journal of Exposure Analysis and Environmental Epidemiology, 23, 13-21.

29.

Pleis

J. R.

Lucas

J. W.

(2007). Summary health statistics for U.S. adults: National Health Interview Survey, 2007. National Center for Health Statistics. Vital and health statistics, 10(240), 1-159.

30.

Pleis

J. R.

Lucas

J. W.

Ward

B. W.

(2009). Summary health statistics for U.S. adults: National Health Interview Survey, 2008. National Center for Health Statistics. Vital and health statistics, 10(242), 1-157.

31.

Priha

Hellman

Sorvari

(2005). PCB contamination from polysulphide sealants in residential areas—exposure and risk assessment. Chemosphere, 59, 537-543.

32.

Ribas-Barba

Serra-Majem

Salvador

Castell

Cabezas

Salleras

Plasencia

(2007). Trends in dietary habits and food consumption in Catalonia, Spain (1992-2003). Public Health Nutrition, 10(11A), 1340-1353.

33.

Schoenborn

C. A.

Adams

P. F.

Barnes

P. M.

Vickerie

J. L.

Schiller

J. S.

(2004). Health behaviors of adults: United States, 1999-2001. National Center for Health Statistics. Vital and health statistics, 10(219), 1-79.

34.

Schoenborn

C. A.

Heyman

K. M.

(2009). Health characteristics of adults aged 55 years and over: United States, 2004-2007 [Report No. 16]. Hyattsville, MD: National Center for Health Statistics.

35.

Sexton

(2012). Cumulative risk assessment: An overview of methodological approaches for evaluating combined health effects from exposure to multiple environmental stressors. International Journal of Environmental Research Public Health, 9, 370-390.

36.

Singleton

J. F.

(1999). Lessons from leisure-time budget research. In Pentland

W. E.

Harvey

A. S.

Lawton

M. P.

McColl

M. A.

(Eds). Time use research in the social sciences (pp. 245-258). New York: Kluwer Academic.

37.

Sonn

(1996). Longitudinal studies of dependence in daily life activities among elderly persons. Scandinavian Journal of Rehabilitation Medicine Supplement, 34, 1-35.

38.

Stalvey

B. T.

Owsley

Sloane

M. E.

Ball

(1999). The life space questionnaire: A measure of the extent of mobility of older adults. Journal of Applied Gerontology, 18, 460-478.

39.

Turczynowicz

Fitzgerald

J. D.

Nitschke

Mangas

McLean

(2007). Site contamination health risk assessment case study involving tenant relocation from a former gasworks site. Journal of Toxicology and Environmental Health Part A, 70, 1638-1653.

40.

Urban

J. D.

Tachovsky

J. A.

Haws

L. C.

Staskal

D. W.

Harris

M. A.

(2009). Assessment of human health risks posed by consumption of fish from the Lower Passaic River, New Jersey. Science of the Total Environment, 408(2), 209-224.

41.

U.S. Environmental Protection Agency. (1997). Exposure factors handbook (Final Report, Report No. EPA/600/P-95/002F a-c). Washington, DC: Author.

42.

U.S. Environmental Protection Agency. (2005a). Guidance on selecting age groups for monitoring and assessing childhood exposures to environmental contaminants (Report No. EPA/630/P-03/003F). Washington, DC: National Center for Environmental Assessment.

43.

U.S. Environmental Protection Agency. (2005b). Aging and toxic response: Issues relevant to risk assessment (Final Report No. EPA/600/P-03/004A). Washington, DC: National Center for Environmental Assessment.

44.

U.S. Environmental Protection Agency. (2007). Summary report of a peer involvement workshop on the development of an Exposure Factors Handbook for the aging (Report No. EPA/600/R-07/061). Washington, DC: National Center for Environmental Assessment.

45.

U.S. Environmental Protection Agency. (2008). Child-specific exposure factors handbook (Final Report, Report No. EPA/600/R-06/096F). Washington, DC: Author.

46.

U.S. Environmental Protection Agency. (2011). Exposure factors handbook: 2011 Edition (Report No. EPA/600/R-090/052F). Washington, DC: Author.

47.

Vadarevu

R. V.

Stopher

P. R.

(1996). Household activities, life cycle, and role allocation. Transportation Research Record, 1556, 77-85.

48.

Veiga

Amaral

Fernandes

H. M.

(1998). Human health risk screening of radioactive and nonradioactive contaminants due to uranium industry operation. Journal of Environmental Radioactivity, 39(1), 69-85.

49.

Wiener

J. M.

Hanley

R. J.

Clark

Van Nostrand

J. F.

(1990). Measuring the activities of daily living: Comparisons across national surveys. Journal of Gerontology, 45(6), S229-S237.