*2.2. Sampling*

This study was conducted in two phases: spring (16 February to 4 April 2014) and summer (20 July to 22 August 2014). Six sites were selected for the study: Kalanki, Balaju, Chabahil, Koteswor, Thapathali, and Jawalakhel. These sites were selected because they were the busiest traffic intersections in the valley. Detailed descriptions of the sampling have previously been published [3,40]. Briefly, at each site, personal exposure of particulate pollution was monitored for six traffic volunteers for up to six days (Sunday to Friday). Sunday is a working day in Nepal. While working on the road, each of the six traffic volunteers carried a bag containing air pollution monitors. These six traffic volunteers worked around the vicinity of each of the six sites. We followed thirty-six traffic volunteers from six sites for six weeks during spring and thirty traffic volunteers from five sites during summer. Volunteers were requested to wear N-95 masks for half of the week (3 days) as an intervention component of the study, where personal exposure to ambient pollutants was dramatically decreased. At each site, all volunteers were requested to wear masks either on the first half of the week (Sunday to Tuesday) or on the second half of the week (Wednesday to Friday). This was done to avoid confounding effects between the use of mask and day of the week. In general, all volunteers have a similar number of working hours (8–10 h), with two working shifts: morning and afternoon. The average temperature during sampling periods in spring and summer was 14.8 ◦C and 23.6 ◦C, respectively; relative humidity in spring and summer was 73.2% and 88.0%, respectively; total precipitations in spring and summer were 50.47 and 266.6 mm, respectively [40].

### *2.3. Blood Sample Collection and Biomarker Analysis*

Blood samples were collected by a professional phlebotomist working at a local hospital. Three blood samples per subject were taken in the beginning, middle, and end of the week separately. In total, there were 78 blood samples from 33 traffic volunteers in the spring season and 63 blood samples from 29 traffic volunteers in the summer season. Blood samples were centrifuged for 15 min within 24 h of sample collection. After centrifugation, serum samples were stored in a freezer at −20 ◦C. Standard deep freezers at –80 ◦C were not available, and are, in fact, rare in Nepal. Samples were then transported under refrigeration to our laboratory at the University of Massachusetts, and then stored at −80 ◦C. No sample quality checks were performed in this set of samples to assess sample degradation.

V-PLEX assay kits (Mesoscale Discovery, Rockville, MD, USA) were used for biomarker analysis. Proinflammatory Panel 1 (human) kits were used for analyzing 10 cytokines: interferon gamma (IFN-γ), interleukins (IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, and IL-13), and tumor necrosis factor (TNF-α) (Table 2). Vascular injury panel 2 (human) kits were used for analyzing serum amyloid A (SAA), CRP, vascular cell adhesion molecule (VCAM-1), and intercellular adhesion molecule (ICAM-1). Protocols from the respective kits were followed to analyze the biomarkers on Discovery Workbench (Mesoscale Discovery, Rockville, MD, USA).


**Table 2.** Comparisons of biomarker concentrations for traffic volunteers in Kathmandu between spring and summer, 2014.

Concentrations: 1 μg/mL; 2 pg/mL; level of significance: \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001. All the tests were two-sided and conducted on the data of each individual's average biomarker concentrations. The independent *t*-test performed a two-sample *t*-test, assuming independence of the volunteer samples in the two seasons. Dependent samples only counted the subjects with biomarker measurements in both seasons. Both a parametric *t*-test and a nonparametric Wilcox test were performed on the dependent samples.
