Intra- and inter-individual variability in location data for two U.S. health-compromised elderly cohorts

Assessing short-term impacts of PM2.5 constituents on cardiorespiratory hospitalizations: Multi-city evidence from China

Apart from concentrations of particulate mass, PM_2.5-associated effects on health may largely depend on its chemical components. However, little is known regarding the underlying effects of specific PM_2.5 constituents. The study included nearly 1 million hospital admissions from five Chinese cities during 2015-2017. Based on the modified Community Multiscale Air Quality model, our study simulated daily concentrations of PM_2.5 and five main components. We used a time-stratified case-crossover design with conditional logistic regression models to estimate short-term effects of PM_2.5 constituents on cause-specific hospital admissions. Per interquartile range increase in exposure to PM_2.5, elemental carbon, organic carbon, nitrate, sulfate and ammonium at lag 04-day was related to an excess risk (ER%) for non-accidental admissions of 1.6% [95% confidence interval: 1.1-2.0], 1.9% [1.3-2.4], 1.0% [0.5-1.6], 1.2% [0.4-2.0], 1.2% [0.9-1.5] and 1.4% [0.9-1.9], respectively. Great heterogeneities of constituents-admission associations existed in diverse causes and constituents. This study provided multi-center high-quality evidence that hospital admissions, particularly those for ischemic heart disease (ER% ranging from 2.3 to 5.4% at lag 04-day) and pneumonia (1.9-5.1% at lag 4-day), could be triggered by short-term exposures to ambient PM_2.5 constituents. Relatively stronger constituents-admission associations were found among females for respiratory causes and the elderly for cardiovascular causes.

Ambient PM2.5 and Risk of Hospital Admissions: Do Risks Differ for Men and Women?

BACKGROUND

While strong evidence exists for associations between fine particles (PM2.5) and health, less is known about whether associations differ by sex.

METHODS

We used Bayesian hierarchical modeling to estimate associations between PM2.5, based on ambient monitors, and risk of cause-specific cardiovascular and respiratory hospitalizations for about 12.6 million Medicare beneficiaries (>65 years) residing in 213 US counties for 1999-2010.

RESULTS

Point estimates were higher for women than men for almost all causes of hospitalization. PM2.5 risks were higher for women than men for respiratory tract infection, cardiovascular, and heart rhythm disturbance admissions. A 10 μg/m(3) increase in same-day PM2.5 was associated with a 1.13% increased risk of heart rhythm disturbance admissions for women (95% posterior interval [PI]: 0.63%, 1.63%), and 0.03% for men (95% PI: -0.48%, 0.55%). Differences remained after stratification by age and season.

CONCLUSIONS

Women may be more susceptible to PM2.5-related hospitalizations for some respiratory and cardiovascular causes.

Statistical properties of longitudinal time-activity data for use in human exposure modeling

Understanding the longitudinal properties of the time spent in different locations and activities is important in characterizing human exposure to pollutants. The results of a four-season longitudinal time-activity diary study in eight working adults are presented, with the goal of improving the parameterization of human activity algorithms in EPA's exposure modeling efforts. Despite the longitudinal, multi-season nature of the study, participant non-compliance with the protocol over time did not play a major role in data collection. The diversity (D)--a ranked intraclass correlation coefficient (ICC)-- and lag-one autocorrelation (A) statistics of study participants are presented for time spent in outdoor, motor vehicle, residential, and other-indoor locations. Day-type (workday versus non-workday, and weekday versus weekend), season, temperature, and gender differences in the time spent in selected locations and activities are described, and D & A statistics are presented. The overall D and ICC values ranged from approximately 0.08-0.26, while the mean population rank A values ranged from approximately 0.19-0.36. These statistics indicate that intra-individual variability exceeds explained inter-individual variability, and low day-to-day correlations among locations. Most exposure models do not address these behavioral characteristics, and thus underestimate population exposure distributions and subsequent health risks associated with environmental exposures.

Estimating lifetime risk from spot biomarker data and intraclass correlation coefficients (ICC)

Human biomarker measurements in tissues including blood, breath, and urine can serve as efficient surrogates for environmental monitoring because a single biological sample integrates personal exposure across all environmental media and uptake pathways. However, biomarkers represent a "snapshot" in time, and risk assessment is generally based on long-term averages. In this study, a statistical approach is proposed for estimating long-term average exposures from distributions of spot biomarker measurements using intraclass correlations based upon measurement variance components from the literature. This methodology was developed and demonstrated using a log-normally distributed data set of urinary OH-pyrene taken from our own studies. The calculations are generalized for any biomarker data set of spot measures such as those from the National Health and Nutrition Evaluation Studies (NHANES) requiring only spreadsheet calculations. A three-tiered approach depending on the availability of metadata was developed for converting any collection of spot biomarkers into an estimated distribution of individual means that can then be compared to a biologically relevant risk level. Examples from a Microsoft Excel-based spreadsheet for calculating estimates of the proportion of the population exceeding a given biomonitoring equivalent level are provided as an appendix.

Longitudinal variability of time-location/activity patterns of population at different ages: a longitudinal study in California

BACKGROUND

Longitudinal time-activity data are important for exposure modeling, since the extent to which short-term time-activity data represent long-term activity patterns is not well understood. This study was designed to evaluate longitudinal variations in human time-activity patterns.

METHOD

We report on 24-hour recall diaries and questionnaires collected via the internet from 151 parents of young children (mostly under age 55), and from 55 older adults of ages 55 and older, for both a weekday and a weekend day every three months over an 18-month period. Parents also provided data for their children. The self-administrated diary and questionnaire distinguished ~30 frequently visited microenvironments and ~20 activities which we selected to represent opportunities for exposure to toxic environmental compounds. Due to the non-normal distribution of time-location/activity data, we employed generalized linear mixed-distribution mixed-effect models to examine intra- and inter-individual variations. Here we describe variation in the likelihood of and time spent engaging in an activity or being in a microenvironment by age group, day-type (weekday/weekend), season (warm/cool), sex, employment status, and over the follow-up period.

RESULTS

As expected, day-type and season influence time spent in many location and activity categories. Longitudinal changes were also observed, e.g., young children slept less with increasing follow-up, transit time increased, and time spent on working and shopping decreased during the study, possibly related to human physiological changes with age and changes in macro-economic factors such as gas prices and the economic recession.

CONCLUSIONS

This study provides valuable new information about time-activity assessed longitudinally in three major age groups and greatly expands our knowledge about intra- and inter-individual variations in time-location/activity patterns. Longitudinal variations beyond weekly and seasonal patterns should be taken into account in simulating long-term time-activity patterns in exposure modeling.