Abstract
The primary aim of this analysis was to present and describe questionnaire data characterizing time-location patterns of an older, multi-ethnic population from six American cities. We evaluated consistency of results from repeated administration of this questionnaire and between this questionnaire and other questionnaires collected from participants of the Multi-Ethnic Study of Atherosclerosis and Air Pollution (MESA Air). Participants reported spending most of their time inside their homes (average: 121 hours/week or 72%). More than 50% of participants reported spending no time in several of the location options, including at home outdoors, at work/volunteer/school locations indoors or outdoors, or in “other” locations outdoors. We observed consistency between self-reported time-location patterns from repeated administration of the time-location questionnaire and compared with other survey instruments. Comparisons to national cohorts demonstrated differences in time-location patterns in the MESA Air cohort due to differences in demographics, but the data showed similar trends in patterns by age, gender, season, and employment status. This study was the first to explicitly examine time-location patterns in an older, multi-ethnic population and the first to add data on Chinese participants. These data can be used to inform future epidemiological research of MESA Air and other studies that include diverse populations.
Keywords: epidemiology, personal exposure, population based studies
Introduction
A standard method for assessing individual-level exposure to environmental pollutants is to calculate the “time-weighted” average of microenvironmental concentrations. This approach can estimate exposure to many different types of pollutants, including air pollutants (1, 2). The more variable the pollutant concentrations are across microenvironments, the greater the need to accurately characterize both the pollutant concentrations and the time spent in each location.
Air pollutant concentrations can be quite variable across proximal microenvironments. Ambient-generated particulate matter (PM) concentrations can decrease 50% or more going from outside to inside a home (3-5). Concentrations of particles, volatile organic compounds, nitric oxide, and nitrogen dioxide inside vehicles traveling on roadways have also been found to be significantly higher than ambient levels (6-13).
Despite this knowledge, most air pollution epidemiology relies on the assumption that residential outdoor concentrations of air pollutants are good proxies for individual exposure (14-21). This assumption is based on data from studies suggesting that individuals spend the majority of time at home (22-25). However, failing to characterize time spent in other microenvironments, especially the indoor environment at home, will likely introduce measurement error. Setton et al. (26) found that residence-only estimates of air pollution exposure have considerable bias compared with estimates that take into account the time spent working or traveling. State-of-the-art air pollution epidemiology is now moving from simply using outdoor concentration data to incorporating information on person-specific characteristics to more accurately estimate individual exposures (27). However, this approach requires more accurate microenvironment time-location data.
Several large-scale studies of time-location patterns provide time-location data for use in exposure assessments. The National Human Activity Pattern Survey (NHAPS) includes 24-hour diary data in 82 different locations for 9,386 participants in 48 states (24). The Consolidated Human Activity Database (CHAD) (22) includes NHAPS, along with data from other national and regional studies. Data from these types of large-scale studies or databases can be a useful alternative to collecting study-specific data. Moreover, these data are used in national exposure models including USEPA air pollution exposure models: Air Pollution Exposure Model (APEX) (28) and Stochastic Human Exposure and Dose Simulation model (SHEDS) (29). However, existing information may not be representative of unique populations. Studies of very specific cohorts (e.g., pregnant women (30); minorities and populations living below the poverty line (31)) have found it necessary to collect data on patterns of time-location specific to their populations. These data not only are valuable for the study in which they were collected, but also broaden the population for which time-location patterns are understood. USEPA has recently made specific recommendations for the collection of additional time-location data for older populations (32).
The primary aim of this analysis is to present the time-location questionnaire data of an older, multi-ethnic cohort of American adults in the Multi-Ethnic Study of Atherosclerosis and Air Pollution (MESA Air) (27). We also evaluate the reliability of time-location data by comparison of repeated administration of the questionnaire and across several survey instruments. Finally, we compare the time-location patterns of MESA Air participants to those in national cohorts. This work adds to the literature as the MESA Air cohort is older and more racially/ethnically diverse than other cohorts in which time-location patterns have previously been described.
Methods
Study Population
MESA Air is an ancillary study to the Multi-Ethnic Study of Atherosclerosis (MESA, (33)), a long-term study of the progression of cardiovascular disease (CVD) in adults. MESA included 6,814 participants from 6 US communities: Baltimore, MD; Chicago, IL; Forsyth County, NC; Los Angeles, CA; New York, NY; and St. Paul, MN. MESA participants were aged 45–84 years at enrollment between 2000 and 2002, with an approximately equal gender ratio, and were free of recognized CVD at baseline. Four ethnic/racial groups were targeted for inclusion: white, black, Hispanic, and Chinese, but recruitment of racial/ethnic groups varied by study site. All MESA Air participants provided informed consent prior to participation.
The primary aim of MESA Air is to understand the relationship between individual-level ambient-derived air pollution exposure and the progression of subclinical CVD and the incidence of clinical disease events (27). MESA Air includes 6,424 participants, primarily recruited between 2005 and 2007 from the MESA cohort. Additional participants were recruited from a second ancillary study to MESA, the MESA Family Study (n=490), and directly for MESA Air (n=257) in three additional areas near existing MESA communities, two in the Los Angeles basin and one near New York City (27). An additional 1,127 MESA participants were included in MESA Air analyses but did not complete all aspects of the MESA Air study and were not included in this analysis.
Overview of Participant Questionnaires
Every MESA Air participant was asked to complete a comprehensive MESA Air Questionnaire (MAQ, see Supplemental Materials) at recruitment (hereafter referred to as the baseline MAQ) to describe typical activity patterns in the winter and summer. These questions were asked during a face-to-face interview during a MESA clinic exam between 2005 and 2007. The MAQ was repeated up to sixtimes during follow up-phone calls and a later clinic exam; repeated administration was triggered for participants who indicated a major change in lifestyle (change in residence, work or school status, caregiver status, or in household members). At a subsequent exam, the MAQ was also re-administered to a small subset of participants (n=10%) who did not report a major lifestyle change to assess changes in responses over time. For participants with multiple MAQs, the time between administration ranged from <3 months to 6.7 years. Detailed daily time location diaries (TLDs, see Supplemental Materials) were also completed by a small subset of participants (n=89) who took part in a personal air pollution exposure study conducted between October 2006 and July 2008 (34). In addition, MESA gathered information on various participant activities including employment and physical activity from all participants via technician-administered Personal History and Physical Activity Questionnaires. Personal History Questionnaires were administered at all clinic exams, while Physical Activity Questionnaires were administered at all but one exam. When the Physical Activity Questionnaire was not administered at the same time as the MAQ, results from the previous clinic exam (230-1,122 days prior to the MAQ) were used. Each of these questionnaires is described in more detail below.
MESA Air Questionnaire (MAQ)
The MAQ was the primary data collection tool for gathering information on home characteristics relevant to pollutant infiltration efficiencies and about behaviors related to individual exposures. Participants were asked specific questions about their typical time-location patterns in winter and summer and could designate if their patterns were the same in both seasons. For each day of the week, the MAQ included questions documenting time spent (rounded to the nearest hour) in each of seven locations: home indoors, home outdoors, work/volunteer/school indoors, work/volunteer/school outdoors, in transit (e.g., car, bike), other indoors, and other outdoors. Participants also designated which days of the week they considered weekends and weekdays. The amount of time by transit mode (e.g., walking/biking, car/taxi, bus, train/subway), road types travelled (e.g., freeways, residential streets), and traffic conditions experienced (e.g., light traffic/moving at the speed limit) were also documented. Additional questions asked about home characteristics related to building type, building age, the presence of an attached garage, and other factors relevant to infiltration. Methods for addressing missing MAQ data are presented in the Supplemental Materials.
Time-Location Diary (TLD)
In the TLDs, the subset of participants undergoing personal sampling recorded hourly time spent in specified locations for each day during two two-week sampling periods using the following classifications: home indoors, home outdoors, motorized vehicle, work indoors, work outdoors, other indoors, and other outdoors. Season classifications were determined using the average temperature over the specific sampling period. If the average temperature was greater than 18°C, the season was classified as “warm”; otherwise, the season was “cool” (3). For comparison to the MAQ, we classified “warm” seasons as summer and “cool” seasons as winter. In the event that we were not able to capture two distinct seasons for an individual participant, results were averaged to obtain a single set of results for each participant for a season. Methods for addressing missing TLD data are presented in the Supplemental Materials.
Personal History and Physical Activity Questionnaires
The Personal History Questionnaire asked all participants about their current employment status in ten job categories: 1) homemaker, 2) employed full time, 3) employed part time, 4) employed – on leave for health reasons, 5) employed – temporarily away from job, 6) unemployed <6 months, 7) unemployed >6 months, 8) retired – not working, 9) retired – working, and 10) retired – volunteering. For comparison to the MAQ, categories 2, 3 and 9 were assigned as working outside of the home, while the remaining seven categories were classified as not working outside of the home. In addition, as part of the Physical Activity Questionnaire, participants provided information on time spent in ten activities: 1) working at their job, 2) conducting volunteer work, 3) walking, 4) doing yard work, 5) traveling via car, subway, or bus, 6) performing household chores, 7) taking care of others, 8) dancing/practicing a sport, 9) doing conditioning activities (e.g., aerobics), and 10) spending time in leisure activities (e.g., reading).
Data Analysis
In order to address our primary aim of presenting time-location data from the MESA Air cohort, we calculated summary statistics for the amount of time spent in each microenvironment based on the MAQ for the cohort as a whole and by demographic group (i.e., age, race/ethnicity, gender, study site, employment status, education, and income, as reported on the Personal History Questionnaire). This cross-sectional analysis included baseline MAQs only.
A second aim of this analysis was to compare time-location patterns obtained from the MAQ to behaviors reported by the same participants on repeated administration of the MAQ and additional survey instruments: the TLD, the Personal History Questionnaire, and the Physical Activity Questionnaire. To accomplish this aim, time-location data for each participant who completed both a TLD and a MAQ in the same season were compared using Spearman correlation coefficients and Wilcoxon Rank Sum tests. For participants who filled out multiple MAQs, we selected the MAQ with the closest date to the TLD. We also investigated the impact of gender, age, race/ethnicity, and employment status on observed differences between the different data collection methods. Consistency across MAQs, for those without lifestyle changes, was assessed with Spearman correlation coefficients. Agreement between the MAQ and the Personal History and Physical Activity Questionnaires was assessed for several measures using Wilcoxon Rank Sum tests. Direct comparisons between the MAQ and the Physical Activity Questionnaire were not made because participants reported time in locations on the MAQ and time in activities on this second survey. Time spent at work was compared among working and non-working persons as classified by the Personal History Questionnaire and among persons reporting time spent at work versus none on the Physical Activity Questionnaire. Time spent outdoors at home was then contrasted between participants reporting some versus no time spent working in the yard on the Physical Activity Questionnaire. Finally, a similar contrast was made for time in transit based on transit and walking behaviors reported in the Physical Activity Questionnaire. All MAQs were matched to Personal History and Physical Activity Questionnaires administered at the same clinical exam or the exam prior for instances when the questionnaire was not administered. Finally, we compared average values of MESA Air time-location data to those reported by NHAPS and CHAD. All statistical analyses were conducted using SAS Software Version 9.3 of the SAS System for Windows (Copyright ©2010 SAS Institute Inc., Cary, NC).
Results
Participant Characteristics
A total of 6,209 participants completed the time-location section of at least one MAQ with 5,996 participants doing so at baseline. Participants completed the MAQ a mean of 1.9 times and a maximum of 6 times for a total of 9,704 MAQs. Of the 9,704 MAQs administered, 7,831 (81%) took place in-person during MESA clinic exams, and 1,873 (19%) occurred during follow-up calls. A total of 5,618 participants at baseline answered the question regarding seasonal differences in time-location patterns, and of these, 45 percent reported the same patterns in winter and summer. Personal History and Physical Activity Questionnaires were only administered at in-clinic exams. For the 7,835 MAQs administered at exams, 7,819 (99.8%) of the Personal History Questionnaires collected at the same exam included responses to the question on job status. For the Physical Activity Questionnaires administered during the same exam visit as the MAQ, 7,680 (98%) of participants provided information on time engaged in various physical activities. Eighty-nine individuals participated in the two-week personal monitoring sampling campaigns and provided at least one TLD. Eighty of these participants participated in a second sampling campaign and completed two TLDs, for a total of 169 TLDs.
Table 1 presents the demographic characteristics of the MESA Air participants who completed baseline MAQs. The average age of participants that completed the baseline MAQ was 65±10 years old, and 53% of the population was female. Fifty percent of the cohort worked full or part time, and another 7% volunteered (data not shown). Demographic characteristics for those that completed the MAQ are very similar to MESA overall (27), and demographic characteristics for participants who completed a baseline MAQ are very similar to those that completed MAQs overall (data not shown).
Table 1.
Characteristics of MESA Air Participants Who Completed MESA Air Questionnaires and Time-Location Diaries
| Baseline MAQs N = 5,996 |
TLDs N = 89 |
|||
|---|---|---|---|---|
| Number | % | Number | % | |
| Gender | ||||
| Female | 3,194 | 53% | 46 | 52% |
| Male | 2,802 | 47% | 43 | 48% |
|
| ||||
| Race/ethnicity | ||||
| White | 2,258 | 38% | 54 | 61% |
| Chinese | 650 | 11% | 2 | 2% |
| Black, African-American | 1,667 | 28% | 23 | 26% |
| Hispanic | 1,421 | 24% | 10 | 11% |
|
| ||||
| Age Category | ||||
| 39-441 | 18 | 0% | 0 | 0% |
| 45-54 | 976 | 16% | 11 | 12% |
| 55-64 | 1,967 | 33% | 37 | 42% |
| 65-74 | 1,815 | 30% | 30 | 34% |
| 75-84 | 1,079 | 18% | 9 | 10% |
| 85+ | 141 | 2% | 2 | 2% |
|
| ||||
| Site | ||||
| Forsyth County | 888 | 15% | 13 | 15% |
| New York City2 | 1,195 | 20% | 21 | 24% |
| Baltimore | 721 | 12% | 14 | 16% |
| St. Paul | 872 | 15% | 13 | 15% |
| Chicago | 1,136 | 19% | 13 | 15% |
| Los Angeles3 | 1,184 | 20% | 15 | 17% |
|
| ||||
| Socioeconomic Status | ||||
| High School Education or More | 4,9994 | 84% | 84 | 94% |
| Family Income ≥$30,000/year | 3,6715 | 65% | 68 | 82% |
| Single Family Home | 3,2666 | 54% | 57 | 64% |
MESA Family included a small number of participants under the age of 45, and 18 completed MAQs
Includes Rockland County participants
Includes Riverside participants
Values missing for 13 participants
Values missing for 340 participants
Values missing for 3 participants
Time-Activity Patterns in MESA Air
On average, participants reported spending the majority (≥70%) of their time indoors at home (Table 2). After time indoors at home, the most time was spent at work/volunteer/school locations indoors; however, more than 50% of participants reported that they spent no time indoors at work/volunteer/school. More than 50% of participants also reported spending no time in the three outdoor locations. Participants reported spending more time in indoor locations and less time in outdoor locations in winter than in summer (Table 2). Figure 1 and Supplemental Tables 2a-g present time-location patterns by age, gender, race/ethnicity, site, employment status, education, and income.
Table 2.
Summary of Reported Time (hours/week, % of total time) by Location Reported by Participants Administered the Baseline MAQ (n=5,996)
| Location | Season | Mean | SD | 5th | 50th | 95th |
|---|---|---|---|---|---|---|
| Indoor Locations | ||||||
|
| ||||||
| Home | Summer | 118 (70%) | 27 (16%) | 75 (45%) | 119 (71%) | 158 (94%) |
| Home | Winter | 125 (74%) | 26 (16%) | 84 (50%) | 126 (75%) | 162 (96%) |
|
| ||||||
| Work/Volunteer/School | Summer | 15 (9%) | 20 (12%) | 0 (0%) | 0 (0%) | 50 (30%) |
| Work/Volunteer/School | Winter | 16 (10%) | 21 (12%) | 0 (0%) | 0 (0%) | 51 (30%) |
|
| ||||||
| Other | Summer | 13 (8%) | 14 (8%) | 0 (0%) | 10 (6%) | 38 (23%) |
| Other | Winter | 14 (8%) | 15 (9%) | 0 (0%) | 11 (6%) | 39 (23%) |
|
| ||||||
| Outdoor Locations | ||||||
|
| ||||||
| Home | Summer | 6 (4%) | 11 (6%) | 0 (0%) | 0 (0%) | 29 (17%) |
| Home | Winter | 2 (1%) | 6 (3%) | 0 (0%) | 0 (0%) | 12 (7%) |
|
| ||||||
| Work/Volunteer/School | Summer | 2 (1%) | 8 (5%) | 0 (0%) | 0 (0%) | 10 (6%) |
| Work/Volunteer/School | Winter | 1 (1%) | 7 (4%) | 0 (0%) | 0 (0%) | 6 (4%) |
|
| ||||||
| Other | Summer | 6 (4%) | 11 (7%) | 0 (0%) | 1 (1%) | 28 (17%) |
| Other | Winter | 2 (1%) | 6 (4%) | 0 (0%) | 0 (0%) | 13 (8%) |
|
| ||||||
| In Transit | Summer | 8 (5%) | 6 (4%) | 0 (0%) | 7 (4%) | 18 (11%) |
| In Transit | Winter | 7 (4%) | 6 (4%) | 0 (0%) | 7 (4%) | 17 (10%) |
Figure 1.
Mean Percent Time Spent in Six Microenvironments by Demographic, Socioeconomic Group, and Site. The remaining time (up to 100 percent) represents the time spent indoors at home. For example, female participants reported spending 74 percent of their time indoors at home.
Consistency of MAQ and TLD Reponses
There were a smaller proportion of Chinese and Hispanics participants in the personal monitoring campaigns, so there are fewer TLDs captured in these groups (Table 1). TLD participants also tended to have higher incomes and were more likely to have at least a high school education. The age distributions of the TLD participants and the cohort as a whole were roughly similar except that TLD participants were more likely to be between the ages of 55 and 64 years and less likely to be 75-84 years. The average time between administration of the MAQ and the start of the TLD was 398 days (range: 13-966 days).
The highest correlations between the TLD and MAQ responses for each individual were for time spent indoors at work and at home (r=0.61 to 0.80), and the weakest correlations were for time spent in other locations (r=−0.04 to 0.28) (Table 3). Agreement between the MAQ and TLD for time spent outdoors was higher in the summer than in the winter. Participants tended to report more time indoors on average (based on their MAQ) than was observed during the two-week personal monitoring period based on their TLDs (data not shown). Percent difference values between the MAQ and TLD for participants who reported on the MAQ having the same time-location patterns in winter and summer were not different from those who reported different seasonal patterns (Wilxon; p>0.05).
Table 3.
Spearman Correlation Coefficients (with 95% confidence intervals) between MAQ and TLD Responses
| Location | Summer (n=65)1 |
Winter (n=78) 1 |
Both (n=143) 1 |
|---|---|---|---|
| Indoor Locations | |||
|
| |||
| Home | 0.61 (0.42, 0.74) |
0.65 (0.49, 0.76) |
0.63 (0.51, 0.72) |
|
| |||
| Work/Volunteer/School | 0.80 (0.68, 0.87) |
0.68 (0.54, 0.78) |
0.73 (0.65, 0.80) |
|
| |||
| Other | 0.25 (0.00, 0.46) |
0.15 (−0.07, 0.36) |
0.20 (0.04, 0.35) |
|
| |||
| All | 0.24 (−0.01, 0.45) |
0.30 (0.08, 0.48) |
0.35 (0.20, 0.49) |
|
| |||
| Outdoor Locations | |||
|
| |||
| Home | 0.19 (−0.06, 0.41) |
0.04 (−0.18, 0.26) |
0.14 (−0.02, 0.30) |
|
| |||
| Work/Volunteer/School | 0.28 (0.03, 0.49) |
0.13 (−0.10, 0.34) |
0.20 (0.03, 0.35) |
|
| |||
| Other | 0.15 (−0.10, 0.38) |
−0.04 (−0.26, 0.18) |
0.10 (−0.06, 0.26) |
|
| |||
| All | 0.08 (−0.17, 0.32) |
0.01 (−0.21, 0.24) |
0.10 (−0.07, 0.26) |
|
| |||
| In Transit | 0.31 (0.06, 0.51) |
0.49 (0.30, 0.64) |
0.39 (0.24, 0.52) |
A total of 89 participants completed TLDs and MAQs. Of these, 65 were in the summer, and 78 were in the winter.
We also examined whether demographic factors affected the strength of the correlations between responses on the MAQ and the TLD. Supplemental Figure 1 shows the Spearman correlation coefficients between the amount of time that each individual reported in each microenvironment on their MAQ and their TLD, by gender, age, race/ethnicity, and employment status. For most locations, men and women provided similarly consistent data on their MAQs and TLDs (Supplemental Figure 1a). However, the responses for the outdoor work location were less consistent for women than men. While none of the women in the personal monitoring campaign reported any time in this location on the MAQ, some women did report spending time at work outdoors on their TLD.
For individuals younger than 65, the two-week periods captured by the TLDs were more highly correlated with the typical patterns of behavior reported in the MAQs than for older individuals (Supplemental Figure 1b). On average, individuals ≥65 years old reported more time indoors and less time outdoors on their MAQ compared to their TLD (data not shown). The correlation of 0 for work outdoors was due to personal monitoring participants ≥65 years old reporting no time on the MAQ and some time on the TLDs. All race/ethnic groups were relatively consistent in their responses regarding time spent in transit. There were differences in the consistency between the MAQ and TLD for the rest of the locations, but these differences were not significant (Supplemental Figure 1c). Working individuals were more consistent in their responses related to work locations but not other locations (Supplemental Figure 1d).
Consistency of Repeated MAQ Administration
A total of 394 participants with no reported lifestyle changes filled out the time-location portion of the MAQ to assess changes in responses over time. The average time between administration of the two MAQs was 1,621 days (range: 22-2,393 days). Correlation between responses for time-location patterns was high (0.78) with higher correlations for indoors (0.89) than transit (0.40) or outdoors (0.35).
Consistency of MAQ Results with other Survey Instruments available in MESA
Comparisons were made between the MAQs and other MESA questionnaires using information on employment status and the time spent at work, working in the yard, in transit, and walking. Not surprisingly, working participants reported spending significantly more time at work compared to individuals in the non-working categories based on the Personal History Questionnaire (means of 30 and 3 hours/week, respectively; p<0.0001) and on the Physical Activity Questionnaire (29 and 4 hours/week; p<0.0001). The amount of time spent outdoors at home from the MAQ was also higher among participants who reported time working in the yard on the Physical Activity Questionnaire than those that did not (6 vs. 2 hours/week; p<0.0001). Participants who reported some transit time on the Physical Activity Questionnaire reported an average of 8 hours per week of transportation time on the MAQ, while those who reported no time in transit on the Physical Activity form reported 5 hours per week on the MAQ (p<0.0001). Participants reporting walking in the Physical Activity Questionnaire also had a higher weekly average amount of time in transit (8 hours) compared with those who did not report walking (6 hours, p<0.0001).
Comparison with Other Studies of Time-Location Patterns
Time-location patterns in the MESA Air cohort were compared with other national cohorts. Comparisons of demographic characteristics for MESA Air, NHAPS, and CHAD are provided in Supplemental Table 3. Although the overall time spent at home indoors was greater for MESA Air than NHAPS participants, similar patterns across seasons and among genders and employment status were observed (Table 4).
Table 4.
Average Time Spent at Home for NHAPS and MESA Air Participants by Season, Gender, and Race (hours/day)
| All | 65+ | Winter | Summer | Male | Female | White | Black | Hispanic | Chinese | Employed full-time |
Employed part-time |
Not Employed |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| time spent indoors at residence | |||||||||||||
|
| |||||||||||||
| NHAPS | 16.7 | 19.6 | 17.2 | 16.3 | 15.8 | 17.5 | 16.7 | 16.9 | 16.8 | NR | 14.7 | 16.4 | 19.3 |
| MESA Air | 17.3 | 18.8 | 17.8 | 16.9 | 16.7 | 17.9 | 17.0 | 17.0 | 17.6 | 18.6 | 14.7 | 17.1 | 19.3 |
|
| |||||||||||||
| time spent outdoors at residence1 | |||||||||||||
|
| |||||||||||||
| NHAPS | 2.3 | 2.4 | 1.9 | 2.4 | 2.6 | 1.9 | 2.3 | 2.1 | 2.1 | NR | 2.2 | 2.1 | 2.4 |
| MESA Air | 1.7 | 1.8 | 1.2 | 2.0 | 1.8 | 1.6 | 1.9 | 1.5 | 1.8 | 1.1 | 1.4 | 1.7 | 1.9 |
NR: not reported
includes only those participants that report some time outdoors at home for both cohorts
Overall, CHAD participants reported spending less time indoors than MESA Air participants. CHAD participants over the age of 65 years spent a similar amount of time indoors and outdoors compared with MESA Air participants over 65 (Table 5). For both CHAD and MESA Air, males reported spending less time indoors and more time outdoors than females, and white participants reported spending the least time indoors and the most outdoors (Table 5).
Table 5.
Comparison of Average Total Time Spent Indoors and Outdoors between CHAD and MESA Air (hours/day)
| All | Male | Female | 65+ | Male: 65+ |
Female: 65+ |
White | Black | Hispanic | Chinese | |
|---|---|---|---|---|---|---|---|---|---|---|
| Total time indoors | ||||||||||
|
| ||||||||||
| CHAD | 21.21 | 20.7 | 21.5 | 21.8 | NR | NR | 21.1 | 21.5 | 21.4 | NR |
| MESA Air | 21.5 | 21.0 | 22.0 | 21.8 | 21.3 | 22.2 | 21.2 | 21.5 | 21.6 | 22.3 |
|
| ||||||||||
| Total time outdoors2 | ||||||||||
|
| ||||||||||
| CHAD | 2.31 | 2.7 | 1.9 | 2.0 | 2.7 | 1.5 | 2.4 | 2.3 | 2.3 | NR |
| MESA Air | 2.4 | 2.8 | 2.0 | 2.3 | 2.7 | 1.9 | 2.7 | 2.5 | 2.3 | 1.4 |
NR: not reported
weighted average of male and female participants
includes only those participants that report some time outdoors for both cohorts
Discussion and Conclusions
In this study, we successfully deployed and evaluated a time-activity questionnaire in a multi-ethnic cohort for incorporation into epidemiological analyses. The MAQ collected information on time spent in microenvironments from MESA Air participants, and agreement between the MAQ and other data collection tools provides evidence of the reliability of this survey.
The evaluation of repeated administration of the MAQ and the comparison of the MAQ to other data collection tools demonstrated consistency across data collection methods, and increases our confidence that the MAQ captured reliable information on where MESA Air participants spend their time. Although variable for location, MAQ and TLD responses were positively correlated. Comparisons between the MAQ and other questionnaires administered in MESA provided additional evidence that the MAQ succeeded in capturing time-location patterns.
Although TLD and MAQ responses were correlated, there were many differences between the MAQ and the TLD. The TLDs represent a two-week snapshot in time, whereas the MAQs are intended to capture typical activities over a season. Requesting information on the “typical” amount of time spent in locations is likely to result in the mode for each participant rather than the mean (35, 36). For participants with less routine schedules, disparity between the MAQ and TLD is more likely. Additionally, because the MESA Air population is older with a large proportion of retired individuals, the time-location pattern for this population is likely less routine than for a working population. If true, we would not necessarily expect any two-week period to be reflective of what is typical but can expect to find similarities in aggregate responses. Overall, we found greater consistency for locations where participants spend more time than for those locations where less time is spent.
Previous researchers found age- and gender-specific differences for the reporting of time spent working (36, 37). Similarly, we found some reporting inconsistencies by gender, age, race/ethnicity, and employment status. Results were similar by gender for all categories except for work outdoors. When we investigated responses by age, we found that agreement between responses to the MAQ regarding typical patterns of behavior and the specific patterns during the two-week period captured by the TLD was lower for most locations for individuals aged ≥65 years. Accuracy of recalled information is likely to decrease as people age, but differences may also be because individuals over the age of 65 are less likely to be working (Wilcoxon, p<0.0001) and therefore may have a less routine schedule.
Differences between MESA Air and NHAPS may be due to differences in demographic composition, collection methodology, and/or dates of collection. In MESA Air, white participants spent the most time outdoors at their residence, and Chinese participants spent the least. Because NHAPS had a large percentage of whites and a very small percentage of Chinese participants (assumed to be some fraction of the 1.7 percent Asian) and a smaller percentage of black and Hispanic participants, we would expect that the MESA Air population would have a lower amount of time outdoors at their residence. NHAPS asked participants to record every minute for a 24-hour period, whereas the MAQ asked participants to record the average time in each location for each season rounded to the nearest hour. Therefore, some differences between NHAPS and MESA Air time-location patterns could be due to this discrepancy in reporting. Because baseline MESA Air data collection occurred 11-15 years after NHAPS, some differences may be due to behavioral changes in the general population over time as the result of changes in technology, weather, socio-economics, and other factors. Participants in MESA Air reported spending more time away from home than those in NHAPS, and this change may be due to changes in lifestyle over this timespan.
MESA Air participants spent more time indoors compared with CHAD participants, which is likely due to differences in the demographics of these two cohorts. For participants at least 65 years of age, time-location patterns were more similar, indicating that the differences between MESA Air and CHAD participants are likely because MESA Air participants are older.
While this study provides reliable estimates of time-location patterns for a diverse, older population, there are some limitations to this analysis. Since the population in MESA is an older population, we cannot use these data to understand time-location in younger groups. The MAQ asked participants to round their time to the nearest hour, so short-term events, even if typical, may not be captured by the MAQ. This approach may result in an underestimation of time spent in certain locations (e.g., home outdoors) and an overestimation of time spent in others (e.g., home indoors). This analysis was limited to a single administration of the MAQ and did not take into account any changes over time. Additionally, too few Chinese participants completed TLDs, so we were not able to assess the MAQ in comparison to the second instrument for this population.
Overall the MAQ has provided useful and reliable estimates of the time spent in multiple microenvironments for the MESA Air cohort, and having information on the amount of time participants spend inside and outside at different locations allows for more precision in air pollution concentration estimates. Future work will assess differences in time-location patterns among demographic groups and investigate how various individual-level characteristics impact total time spent in specific microenvironments.
Supplementary Material
Acknowledgements
This publication was developed under a STAR research assistance agreement, No. RD831697 (MESA Air) and grant No. RD-83479601-0 (CCAR) by the U.S. Environmental Protection Agency. It has not been formally reviewed by the EPA. The views expressed in this document are solely those of the authors and the EPA does not endorse any products or commercial services mentioned in this publication. Support for MESA is provided by contracts N01-HC-95159, N01-HC-95160, N01-HC-95161, N01-HC-95162, N01-HC-95163, N01-HC-95164, N01-HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168, and N01-HC-95169 and CTSA UL1-RR-024156. Funding for the MESA Family study is provided by grants R01-HL-071051, R01-HL-071205, R01-HL-071250, R01-HL-071251, R01-HL-071252, R01-HL-071258, R01-HL-071259, UL1-RR-025005 by the National Center for Research Resources, Grant UL1RR033176, and is now at the National Center for Advanced Translational Sciences, Grant UL1TR000124. Additional support was provided from the National Institute of Environmental Health Sciences through grants K24ES013195, P50ES015915, and P30ES07033.
References
- 1.NRC. Human Exposure Assessment for Airborne Pollutants: Advances and Opportunities. National Research Council, The National Academies Press; Washington, DC: 1991. [Google Scholar]
- 2.Sexton K, Ryan PB. In: Assessment of human exposure to air pollution: methods, measurements, and models. Watson AY, Bates RR, Kennedy D, editors. National Academy Press; Air Pollution, the Automobile, and Public Health New York: 1988. pp. 207–38. [PubMed] [Google Scholar]
- 3.Allen RW, Adar SD, Avol E, Cohen M, Curl CL, Larson T, et al. Modeling the residential infiltration of outdoor PM(2.5) in the Multi-Ethnic Study of Atherosclerosis and Air Pollution (MESA Air) Environmental health perspectives. 2012;120(6):824–30. doi: 10.1289/ehp.1104447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chen C, Zhao B. Review of relationship between indoor and outdoor particles: I/O ratio, infiltration factor and penetration factor. Atmospheric Environment. 2011;45(2):275–88. [Google Scholar]
- 5.Allen R, Larson T, Sheppard L, Wallace L, Liu L-JS. Use of real-time light scattering data to estimate the contribution of infiltrated and indoor-generated particles to indoor air. Environmental science & technology. 2003;37(16):3484–92. doi: 10.1021/es021007e. [DOI] [PubMed] [Google Scholar]
- 6.Jo W-K, Park K-H. Concentrations of volatile organic compounds in the passenger side and the back seat of automobiles. Journal of Exposure Analysis and Environmental Epidemiology. 1999;9(3):217. doi: 10.1038/sj.jea.7500041. [DOI] [PubMed] [Google Scholar]
- 7.Leung P-L, Harrison RM. Roadside and in-vehicle concentrations of monoaromatic hydrocarbons. Atmospheric environment. 1999;33(2):191–204. [Google Scholar]
- 8.Westerdahl D, Fruin S, Sax T, Fine PM, Sioutas C. Mobile platform measurements of ultrafine particles and associated pollutant concentrations on freeways and residential streets in Los Angeles. Atmospheric Environment. 2005;39(20):3597–610. [Google Scholar]
- 9.Chan CC, Ozkaynak H, Spengler JD, Sheldon L. Driver exposure to volatile organic compounds, carbon monoxide, ozone and nitrogen dioxide under different driving conditions. Environmental science & technology. 1991;25(5):964–72. [Google Scholar]
- 10.Adar SD, Davey M, Sullivan JR, Compher M, Szpiro A, Liu LJ. Predicting Airborne Particle Levels Aboard Washington State School Buses. Atmospheric Environment. 2008;42(33):7590–9. doi: 10.1016/j.atmosenv.2008.06.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Adar SD, Gold DR, Coull BA, Schwartz J, Stone PH, Suh H. Focused exposures to airborne traffic particles and heart rate variability in the elderly. Epidemiology. 2007;18(1):95–103. doi: 10.1097/01.ede.0000249409.81050.46. [DOI] [PubMed] [Google Scholar]
- 12.Behrentz E, Sabin LD, Winer AM, Fitz DR, Pankratz DV, Colome SD, et al. Relative importance of school bus-related microenvironments to children’s pollutant exposure. Journal of the Air & Waste Management Association. 2005;55(10):1418–30. doi: 10.1080/10473289.2005.10464739. [DOI] [PubMed] [Google Scholar]
- 13.Zhang KM, Wexler AS, Zhu YF, Hinds WC, Sioutas C. Evolution of particle number distribution near roadways. Part II: the ‘Road-to-Ambient’process. Atmospheric Environment. 2004;38(38):6655–65. [Google Scholar]
- 14.Dockery DW, Pope CA, 3rd, Xu X, Spengler JD, Ware JH, Fay ME, et al. An association between air pollution and mortality in six U.S. cities. The New England journal of medicine. 1993;329(24):1753–9. doi: 10.1056/NEJM199312093292401. [DOI] [PubMed] [Google Scholar]
- 15.Miller KA, Siscovick DS, Sheppard L, Shepherd K, Sullivan JH, Anderson GL, et al. Long-term exposure to air pollution and incidence of cardiovascular events in women. The New England journal of medicine. 2007;356(5):447–58. doi: 10.1056/NEJMoa054409. [DOI] [PubMed] [Google Scholar]
- 16.Pope CA, 3rd, Burnett RT, Thun MJ, Calle EE, Krewski D, Ito K, et al. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA : the journal of the American Medical Association. 2002;287(9):1132–41. doi: 10.1001/jama.287.9.1132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hoek G, Brunekreef B, Goldbohm S, Fischer P, van den Brandt PA. Association between mortality and indicators of traffic-related air pollution in the Netherlands: a cohort study. The Lancet. 2002;360(9341):1203–9. doi: 10.1016/S0140-6736(02)11280-3. [DOI] [PubMed] [Google Scholar]
- 18.Ostro B, Lipsett M, Reynolds P, Goldberg D, Hertz A, Garcia C, et al. Long-term exposure to constituents of fine particulate air pollution and mortality: results from the California Teachers Study. Environmental health perspectives. 2010;118(3):363–9. doi: 10.1289/ehp.0901181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Gehring U, Heinrich J, Krämer U, Grote V, Hochadel M, Sugiri D, et al. Long-term exposure to ambient air pollution and cardiopulmonary mortality inwomen. Epidemiology. 2006;17(5):545–51. doi: 10.1097/01.ede.0000224541.38258.87. [DOI] [PubMed] [Google Scholar]
- 20.Rosenlund M, Berglind N, Pershagen G, Hallqvist J, Jonson T, Bellander T. Long-term exposure to urban air pollution and myocardial infarction. Epidemiology. 2006;17(4):383–90. doi: 10.1097/01.ede.0000219722.25569.0f. [DOI] [PubMed] [Google Scholar]
- 21.Puett RC, Hart JE, Yanosky JD, Paciorek C, Schwartz J, Suh H, et al. Chronic fine and coarse particulate exposure, mortality, and coronary heart disease in the Nurses' Health Study. Environmental health perspectives. 2009;117(11):1697–701. doi: 10.1289/ehp.0900572. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Graham SE, McCurdy T. Developing meaningful cohorts for human exposure models. Journal of Exposure Science and Environmental Epidemiology. 2004;14(1):23–43. doi: 10.1038/sj.jea.7500293. [DOI] [PubMed] [Google Scholar]
- 23.U.S EPA. EPA Report. 2011. Exposure factors handbook 2011 edition (final) EPA/600/R-09/052F. [Google Scholar]
- 24.U.S EPA. Descriptive Statistics Tables from a Detailed Analysis of the National Human Activity Pattern Survey (NHAPS) Data. Environmental Protection Agency;; Las Vegas, NV: U.S.: 1996. [Google Scholar]
- 25.Leech JA, Nelson WC, Burnett RT, Aaron S, Raizenne ME. It's about time: a comparison of Canadian and American time-activity patterns. Journal of Exposure Analysis and Environmental Epidemiology. 2002;12(6):427–32. doi: 10.1038/sj.jea.7500244. [DOI] [PubMed] [Google Scholar]
- 26.Setton E, Marshall JD, Brauer M, Lundquist KR, Hystad P, Keller P, et al. The impact of daily mobility on exposure to traffic-related air pollution and health effect estimates. Journal of exposure science & environmental epidemiology. 2011;21(1):42–8. doi: 10.1038/jes.2010.14. [DOI] [PubMed] [Google Scholar]
- 27.Kaufman JD, Adar SD, Allen RW, Barr RG, Budoff MJ, Burke GL, et al. Prospective study of particulate air pollution exposures, subclinical atherosclerosis, and clinical cardiovascular disease: The Multi-Ethnic Study of Atherosclerosis and Air Pollution (MESA Air) American journal of epidemiology. 2012;176(9):825–37. doi: 10.1093/aje/kws169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.U.S EPA. Environmental Protection Agency; Research Triangle Park, NC: U.S.: 2012. otal Risk Integrated Methodology (TRIM) Air Pollutants Exposure Model Documentation (TRIM.Expo/APEX, Version 4.5), Volume II: Technical Support Document 2012 US EPA Office of Air Quality Planning and Standards. [Google Scholar]
- 29.U.S EPA. Stochastic Human Exposure and Dose Simulation Model for Multimedia, Multipathway Chemicals, SHEDS-Multimedia Model Version 3 Technical Manual. Environmental Protection Agency; Washington, DC: U.S.: 2008. [Google Scholar]
- 30.Nethery E, Leckie SE, Teschke K, Brauer M. From measures to models: an evaluation of air pollution exposure assessment for epidemiological studies of pregnant women. Occupational and environmental medicine. 2008;65(9):579–86. doi: 10.1136/oem.2007.035337. [DOI] [PubMed] [Google Scholar]
- 31.Wu XM, Fan ZT, Ohman-Strickland P. Time-location patterns of a population living in an air pollution hotspot. Journal of Environmental and Public Health. 2010;2010 doi: 10.1155/2010/625461. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.U.S EPA. Data Sources Available for Modeling Environmental Exposures in Older Adults. EPA Report. 2011 EPA/600/R-12/013. [Google Scholar]
- 33.Bild DE, Bluemke DA, Burke GL, Detrano R, Diez Roux AV, Folsom AR, et al. Multi-ethnic study of atherosclerosis: objectives and design. American journal of epidemiology. 2002;156(9):871–81. doi: 10.1093/aje/kwf113. [DOI] [PubMed] [Google Scholar]
- 34.Cohen MA, Adar SD, Allen RW, Avol E, Curl CL, Gould T, et al. Approach to estimating participant pollutant exposures in the Multi-Ethnic Study of Atherosclerosis and Air Pollution (MESA Air) Environmental science & technology. 2009;43(13):4687–93. doi: 10.1021/es8030837. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Lin K-H. Revisiting the gap between stylized and diary estimates of market work time. Social Science Research. 2012;41(2):380–91. doi: 10.1016/j.ssresearch.2011.11.013. [DOI] [PubMed] [Google Scholar]
- 36.Juster FT, Ono H, Stafford FP. An assessment of alternative measures of time use. Sociological Methodology. 2003;33(1):19–54. [Google Scholar]
- 37.Bonke J. Paid Work and Unpaid Work: Diary Information Versus Questionnaire Information. Soc Indic Res. 2005;70(3):349–68. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.