TY - JOUR
AU - Hensel, Edward, C
AB - Abstract Introduction Little is known about the natural use behavior of new and emerging tobacco products due to the limited availability of reliable puff topography monitors suitable for ambulatory deployment. An understanding of use behavior is needed to assess the health impact of emerging tobacco products and inform realistic standardized topography profiles for emissions studies. The purpose of this study is to validate four monitors: the wPUM cigalike, vapepen, hookah, and cigarette monitors. Aims and Methods Each wPUM monitor was characterized and validated for range, accuracy, and resolution for puff flow rate, duration, volume, and interpuff gap in a controlled laboratory environment. Monitor repeatability was assessed for each wPUM monitor using four separate week-long natural environment monitoring studies including cigalike, vapepen, hookah, and cigarette users. Results The valid flow rate range was 10 to 100 mL/s for cigalike and cigarette monitors, 10 to 95 mL/s for vapepen monitors, and 50 to 400 mL/s hookah monitors. Flow rate accuracy was within ±2 mL/s for cigalike, vapepen, and cigarette monitors and ±6 mL/s for the hookah monitor. Durations and interpuff gaps as small as 0.2 s were measured to within ±0.07 s. Monitor calibrations changed by 4.7% (vapepen), 1.5% (cigarette), 0.5% (cigalike), and 0.1% (hookah) after 1 week of natural environment use. Conclusions The wPUM topography monitors were demonstrated to be reliable when deployed in the natural environment for a range of emerging tobacco products. Implications The current study addresses the lack of available techniques to reliably monitor topography in the natural environment, across multiple emerging tobacco products. Natural environment topography data will inform standardized puffing protocols for premarket tobacco product applications. The ability to quantify topography over extended periods of time will lead to a better understanding of use behavior and better-informed regulations to protect public health. Introduction An understanding of tobacco use behavior outside the lab environment is needed to assess the health impact of emerging tobacco products and inform realistic standardized topography profiles for emissions studies. Ongoing efforts to observe realistic puffing topography of various inhaled tobacco products, including combustible cigarettes, electronic cigarettes (ecigs), cigars, and hookahs, are hampered by the limited availability of reliable monitoring instruments and techniques. Assessing topography in a laboratory environment, although well controlled, may not accurately represent real-word use.1 In addition, related biological effects of tobacco product use may be altered in laboratory situations.2 Even in the laboratory setting, limitations of commercially available topography monitors have been identified. For example, the CReSS reportedly suffers from device failure causing loss of data,3 inaccurate puff counts and maximum recordable puff count of 43 puffs,4 maximum operating limit of 20 min, and maximum puff duration of 5 s,5 and lack of puff recognition below 20 mL/s.6 Assessing topography in the natural environment (NE), although more representative of actual use, has the added uncertainty of whether or not the monitor can sustain its performance over time in the field.5 Comprehensive validation data for NE monitors are lacking and are needed to provide context and uncertainty bounds on reported topography data and the impact of topography on yield.7 Puffing topography is quantified by puff flow rate, puff duration, puff volume, interpuff interval, and puff count. Various techniques have been employed including live video recording,8 data mining internet videos,9 hand gestures,10 and commercial and noncommercial topography monitors.3, 4, 11–18 Video recordings and hand gestures are limiting because only puff duration, interpuff interval and puff count can be measured, whereas flow rate and volume are needed to assess emissions.19 Some ecigs such as eVIC can report puff duration and have been used in NE topography studies.20 Few options are available for capturing all aspects of puff topography, and these options are limited either by performance uncertainty or lack of validation in the NE. A recent paper provided the first comprehensive validation study of two commercially available monitors, the CReSS and the SPA-M, and the monitors’ ability to capture ecig puffing behavior in controlled laboratory machine tests and real-world settings.5 The study reported recording anomalies from a 1-week NE study such as inaccurate detection of the end of a puff, flow rate dropping to zero within a puff (13–49% of the time for SPA and 40–100% for CReSS), unreported puffs, and maximum vape session durations of 20 min or 2000 samples for the CReSS, and multiple peaks being counted as one puff for the SPA-M. For controlled in-lab tests, the study reported puff volume and duration accuracies of about 10% for the SPA-M depending on the range, but widely varying accuracies for the CReSS. Researchers at the British American Tobacco Company presented validation data for a research-grade laboratory topography monitor that was adapted for use with ecigs and designed to overcome some of the previously reported limitations: “indirect determination of puff durations, deposition of aerosol within the topography device affecting volume measurements and reliable measurements at low flow rates.” The study reports the device was validated in a laboratory setting with aerosol volumes reportedly being within 5 mL of the target volume.21 The monitor requires computer connectivity to operate and was not described as portable or suitable for ambulatory studies. There is a need for validated techniques to reliably monitor topography in the NE. The purpose of the current study was to address the lack of available techniques to reliably monitor topography in the NE for new and emerging tobacco products. The wPUM monitors have been previously deployed in the NE, but a comprehensive characterization and validation study has not yet been presented. The wPUM monitors have four different form factors each specific to a product type (cigalike, vapepen, hookah, and cigarette), while utilizing common hardware and software components. The study presents validation data focused on the monitors’ application to emerging tobacco products and natural use behavior. First, this study reports accuracy, operating range, and resolution for all topography puffing parameters, including puff flow rate, duration, interpuff gap and volume based on well-controlled in-lab experiments. Second, the study reports the repeatability of monitors that were deployed for 1-week in the NE. Materials and Methods wPUM Topography Monitors and TAP Analysis Program The performance of four wPUM NE topography monitors, previously described, was evaluated in this study; wPUM Cigalike Monitor,16,17 wPUM Vapepen Monitor,18 wPUM Hookah Monitor,22 and wPUM Cigarette Monitor. Photos are provided in supplemental images (Supplementary Figure S1). The wPUM monitors were developed by engineers in the Respiratory Technologies Laboratory (RTL) at Rochester Institute of Technology (RIT). These monitors are referred to as the second-generation family which share a common electrical design, including the core board, pressure sensor, and firmware distinct from that of the first generation wPUM monitors previously presented.16 These common the second-generation electronics were housed in cases unique to the tobacco product being tested. The form factor of the case was designed to be minimally intrusive so as not to drastically alter how the user would naturally hold the product. Likewise, the shape of the mouthpiece on the monitor was nearly identical to the mouthpiece of the product. The cigalike and cigarette monitor cases were identical except that the cigalike monitor had a removable adapter at the distal end of the mouthpiece to hold a cigalike in place. The wPUM monitors were portable and self-contained. They did not require connection to a computer or any other ancillary device and had the capability to collect at least 1 week of data on a fresh set of batteries. The units were powered by the user via an on/off switch and continuously collected data to a microSD card at a rate of 40 Hz while powered on. The on-board real-time clock provided a date and time stamp for each data point. Data were transferred to the laboratory computer by the research administrator via microSD card. The flow rate was derived from the direct measurement of pressure drop across a calibrated orifice plate, via the TAP topography analysis software, which is a suite of data processing algorithms developed by RTL engineers at RIT. By running the TAP code, noisy raw voltage data were converted to a puff profile, or flow rate as a function of time. Next puff duration was determined by identifying the start and end of each puff using the analysist’s predefined minimum flow rate criteria associated with the signal-to-noise ratio of the monitor. Finally, puff volume was determined by integrating under the puff profile for each identified puff. The TAP software had the capability to produce descriptive topography statistics by puff, session, subject, or study cohort, and compute inferential statistics to determine significant differences between groups. Programmable Emissions Systems Puff profiles were generated with one of three different RIT programmable emission systems (PES), which were designed by RTL engineers at RIT and used previously.23–25 Each PES was operated by a closed-loop control algorithm operating at about 20 Hz and utilized a calibrated flow meter manufactured by Alicat Scientific, Inc. The difference between the target puff profiles (PES Command) and the actual puff profiles (PES Observed) was determined for each test to confirm the desired testing conditions were met. The TAP code computed the topography parameters from the Alicat flow rate readings taken every 0.05 s. PES-0, used to characterize the cigarette monitors, and the PES-1, used to characterize the cigalike and vapepen monitors, generated puffs from 10 to 150 mL/s with accuracy within ±2 mL/s or within 1.8 % of command. PES-2, used to characterize the hookah monitor, generated puffs from 50 to 500 mL/s with accuracy within ±5 mL/s or 1.4% of command. The PES machines generated durations and interpuff gaps from 0.2 to 10 s with accuracy within ±0.1 s. The PES Observed values served as the reference for monitor calibration and monitor validation. Monitor Calibration Alicat flow meters served as the reference instruments for calibrating the wPUM monitors. The Alicat flow meters were calibrated annually by the manufacturer. The PES machines were used to generate calibration puff profiles consisting of 5 s duration square wave puffs with varying flow rates. For the cigalike, vapepen, and cigarette monitors, the calibration puff profiles included 20 different puff flow rates varying from 10 to 100 mL/s. For the hookah monitor, the calibration puff profiles included 15 different puff flow rates varying from 50 to 400 mL/s. The calibration coefficients were derived using a regression model of the form q = k (V – V0)1/2, where q was the Alicat or PES Observed flow rate (in mL/s), V and V0 were the wPUM monitor’s pressure sensor voltage response (in counts) during flow and no flow conditions, respectively, and k was the calibration coefficient (in mL/s/count1/2). Calibration coefficients were determined before (precalibration k) and after (postcalibration k) each in-lab trial and each NE deployment. In-Lab Validation The wPUM monitors were exposed to three repeated trials each of three different square wave puff profiles; vary puff flow rate (profile 1), vary puff duration (profile 2), and vary interpuff gap (profile 3). Square wave profiles were used because they approximate the shape of the puffing profiles observed in the field16 and because square waves are the most difficult puff shape to capture due to high sampling frequency requirements. Profile 1 consisted of 50 puffs with varying flow rates from 10 to 150 mL/s for the cigalike, vapepen, and cigarette monitors, and 34 puffs with varying flow rates from 50 to 500 mL/s for the hookah monitor. For profile 1, the puff duration was 10 s, and the interpuff gap was 20 s. Profile 2 consisted of 50 puffs with varying durations from 10 s down to 0.2 s, with interpuff gap of 20 s. For cigalike, vapepen, and cigarette monitors, Profile 2 had 30 mL/s puffs, and for the hookah monitor Profile 2 had 225 mL/s puffs. Profile 3 consisted of 50 puffs with varying interpuff gap from 10 s down to 0.2 s, with puff durations of 10 s. For cigalike, vapepen, and cigarette monitors, Profile 3 had 30 mL/s puffs, and for the hookah monitor Profile 3 had 225 mL/s puffs. The data file used to drive the PES profile provides the Command flow rate at intervals of 0.01 s, well below the closed-loop feedback control loop period (0.05 s) of the PES. The flow conditions observed by the Alicat flow meters, or PES Observed, were used to compare wPUM measurements, referred to as wPUM Reading. The TAP code was used to compute the puff-by-puff topography parameters from PES Observed data and from wPUM Reading data. Puff counts detected by the wPUM monitors were compared with the puff counts presented in the three test profiles. Between the three profiles, a total of 390 puffs times 3 repeated trials were tested for the cigalike, vapepen, and cigarette, and a total of 402 puffs times 3 repeated trials was tested for hookah. The test profiles spanned the monitor’s design operating range. That is, our goal was to demonstrate the monitors could detect puffs as short as 0.2 s, at flow rates as low as 10 mL/s for cigalike, vapepen, and 50 mL/s for hookah and recognize separate individual puffs spaced as close as 0.2 s apart. Puff counts were assessed for each different puff and for the 1572 total puffs. Topography parameters for each puff were compared between the wPUM Reading data and the PES Observed data on a puff-by-puff basis. Error was computed as Error = wPUM Reading – PES Observed. Accuracy was conservatively estimated to be plus or minus the absolute value of the maximum error. Since puff volume was not measured directly, but rather computed from the measured puff flow rate and duration, puff volume accuracy was assessed using error propagation theory to account for measurement accuracies of puff flow rate and duration. Specifically, Volume Error = Flow Rate Error + Duration Error + Flow Rate Error * Duration Error. The validated operating range for puff flow rate and duration was taken by visual inspection to be the point beyond which the wPUM monitor was no longer able to respond to changes in the input profile, which was an indication the upper limit of the pressure sensor was encountered. The validated operating range for puff volume was taken as the product of puff flow rate and duration at the upper and lower ends of the validated range for puff flow rate and duration. The resolution of flow rate measured via an orifice plate increases with increasing flow rate according to a power law relationship and is dependent upon the resolution of the analog-to-digital (A/D) converter employed. Resolution was determined by applying the power law relationship, q = k(V – V0)1/2 for single bit changes in the analog-to-digital converter at the upper and lower end of the validated flow rate range, accounting for the 10 bit resolution of the analog-to-digital converter. At the upper end, V–V0 was 512 voltage counts, giving a maximum flow rate resolution = k [(512)1/2 – (511)1/2]. Similarly, the minimum flow rate resolution = k [(1)1/2 – (0)1/2]. Resolution for each wPUM was then determined using the average of the pre- and postcalibration coefficients. NE Validation Topography monitor performance across four previously conducted natural environment studies was assessed. The first study17 involved cigalike users, the wPUM Cigalike Monitor, and 20 participants using a variety of own-choice cigalike products including Blu, Logic, Zoom, V2, Markten, NJOY, and Criss Cross. The second study18 involved vapepen users, the wPUM Vapepen Monitor, and 33 participants using NJOY “VapePen” prefilled with Avail brand “Tobacco Row” flavored eliquid. The third study22 involved hookah users, the wPUM Hookah Monitor, and 24 participants using their own hookah and choice shisha. The fourth study involved cigarette users, the wPUM Cigarette Monitor, and 26 participants using variety of own-choice products, including various types of Marlboro, Camel, and Seneca brands. In each study, experienced users took the product-specific wPUM topography monitors home and were instructed to continue using their tobacco product ad lib with the wPUM monitors for 1 week. To assess the measurement repeatability of the wPUM monitors, the pre- and postcalibration intraclass correlation coefficient (ICC) was determined for each type of monitor. Results Range and Accuracy Results of the in-lab validation test for accuracy in puff count, flow rate, duration, interpuff gap, and volume across the validated flow rate ranges are summarized in Table 1. Puff count accuracy across the three repeated trials for all four monitors was 100%; 1572 puffs presented to the monitors were detected. Results confirmed the monitors can detect puffs as short as 0.2 s, separated by at least 0.2 s having flow rates as low as 10 mL/s for cigalike, vapepen, and hookah and 50 mL/s for hookah. No spurious puffs were recorded. As expected, the flow rate performance of the cigalike, vapepen, and cigarette monitors was comparable. These three monitors were designed for low flow rates and achieved an operating range of 10 to 95 mL/s within ±2 mL/s error. The flow rate error for hookah monitor, which was designed to capture higher flow rates, was ±6 mL/s across the validated operating range of 50 to 400 mL/s. For readability, scatter plots are shown in Figure 1 for cigalike, vapepen, and cigarette monitors and Supplementary Figure S2 for hookah monitors. The four monitor designs were expected to perform similarly relative to duration accuracy and interpuff gap accuracy, since the monitors share common hardware and sampling frequency. All four monitors performed within an error of ±0.1 s for both duration and interpuff gap across the range tested. The relative error in the time domain starts at 5% for a typical 2 s puff and decreases to 1% for a 10-s puff. Note that there is no upper limit on the puff duration that can be recorded. Exemplar scatter plots are shown in Figures 2 and 3 for duration and interpuff gap error, respectively. Table 1. Results for range, accuracy, resolution and repeatability for each monitor design . wPUM Gen 2 Cigalike . wPUM Gen 2 Vapepen . wPUM Gen 2 Hookah . wPUM Gen 2 cigarette . Validated operating range1 Puff flow rate, mL/s 10 to 95 10 to 95 50 to 400 10 to 95 Puff duration2, s 0.2 to 10 0.2 to 10 0.2 to 10 0.2 to 10 Puff gap2, s 0.2 to 10 0.2 to 10 0.2 to 10 0.2 to 10 Puff volume, mL2 6 to 950 6 to 950 45 to 4,000 6 to 950 Accuracy Puff flow rate, mL/s ± 2 ± 2 ± 30 ± 2 Puff duration, s ± 0.1 ± 0.1 ± 0.1 ± 0.1 Puff gap, s ± 0.1 ± 0.1 ± 0.1 ± 0.1 Puff volume, mL ± 2.3 ± 2.3 ± 33 ± 2.3 Resolution Puff flow rate, mL/s 0.1 to 5.0 0.1 to 5.0 0.4 to 20.0 0.1 to 5.0 Puff duration, s 0.025 0.025 0.025 0.025 Puff gap, s 0.025 0.025 0.025 0.025 Puff volume, mL 0.02 to 1.00 0.02 to 1.00 0.08 to 4.00 0.02 to 1.00 Repeatability Number of monitors and deployments 6 deployments across 3 monitors 12 deployments across 8 monitors 16 deployments across 4 monitors 18 deployments across 6 monitors In-Lab PES/wPUM predeployment calibration ICC1:11 3/3 ≥ 0.97 8/8 ≥ 0.98 4/4 ≥ 0.99 6/6 ≥ 0.99 Natural environment pre/postdeployment calibration ICC1:11 3/3 ≥ 0.94 6/8 ≥ 0.90 7/8 ≥ 0.83 for all but one deployment of one monitor3 1/8 = 0.35 4/4 ≥ 0.96 6/6 ≥ 0.98 for all but one deployment of one monitor4 . wPUM Gen 2 Cigalike . wPUM Gen 2 Vapepen . wPUM Gen 2 Hookah . wPUM Gen 2 cigarette . Validated operating range1 Puff flow rate, mL/s 10 to 95 10 to 95 50 to 400 10 to 95 Puff duration2, s 0.2 to 10 0.2 to 10 0.2 to 10 0.2 to 10 Puff gap2, s 0.2 to 10 0.2 to 10 0.2 to 10 0.2 to 10 Puff volume, mL2 6 to 950 6 to 950 45 to 4,000 6 to 950 Accuracy Puff flow rate, mL/s ± 2 ± 2 ± 30 ± 2 Puff duration, s ± 0.1 ± 0.1 ± 0.1 ± 0.1 Puff gap, s ± 0.1 ± 0.1 ± 0.1 ± 0.1 Puff volume, mL ± 2.3 ± 2.3 ± 33 ± 2.3 Resolution Puff flow rate, mL/s 0.1 to 5.0 0.1 to 5.0 0.4 to 20.0 0.1 to 5.0 Puff duration, s 0.025 0.025 0.025 0.025 Puff gap, s 0.025 0.025 0.025 0.025 Puff volume, mL 0.02 to 1.00 0.02 to 1.00 0.08 to 4.00 0.02 to 1.00 Repeatability Number of monitors and deployments 6 deployments across 3 monitors 12 deployments across 8 monitors 16 deployments across 4 monitors 18 deployments across 6 monitors In-Lab PES/wPUM predeployment calibration ICC1:11 3/3 ≥ 0.97 8/8 ≥ 0.98 4/4 ≥ 0.99 6/6 ≥ 0.99 Natural environment pre/postdeployment calibration ICC1:11 3/3 ≥ 0.94 6/8 ≥ 0.90 7/8 ≥ 0.83 for all but one deployment of one monitor3 1/8 = 0.35 4/4 ≥ 0.96 6/6 ≥ 0.98 for all but one deployment of one monitor4 1. There is no upper limit to this parameter. 2For all deployments of each monitor unless noted. 3One monitor had ICC1:1 = 0.18 for one deployment. 4One monitor had ICC1:1 = 0.63 for one deployment. Open in new tab Table 1. Results for range, accuracy, resolution and repeatability for each monitor design . wPUM Gen 2 Cigalike . wPUM Gen 2 Vapepen . wPUM Gen 2 Hookah . wPUM Gen 2 cigarette . Validated operating range1 Puff flow rate, mL/s 10 to 95 10 to 95 50 to 400 10 to 95 Puff duration2, s 0.2 to 10 0.2 to 10 0.2 to 10 0.2 to 10 Puff gap2, s 0.2 to 10 0.2 to 10 0.2 to 10 0.2 to 10 Puff volume, mL2 6 to 950 6 to 950 45 to 4,000 6 to 950 Accuracy Puff flow rate, mL/s ± 2 ± 2 ± 30 ± 2 Puff duration, s ± 0.1 ± 0.1 ± 0.1 ± 0.1 Puff gap, s ± 0.1 ± 0.1 ± 0.1 ± 0.1 Puff volume, mL ± 2.3 ± 2.3 ± 33 ± 2.3 Resolution Puff flow rate, mL/s 0.1 to 5.0 0.1 to 5.0 0.4 to 20.0 0.1 to 5.0 Puff duration, s 0.025 0.025 0.025 0.025 Puff gap, s 0.025 0.025 0.025 0.025 Puff volume, mL 0.02 to 1.00 0.02 to 1.00 0.08 to 4.00 0.02 to 1.00 Repeatability Number of monitors and deployments 6 deployments across 3 monitors 12 deployments across 8 monitors 16 deployments across 4 monitors 18 deployments across 6 monitors In-Lab PES/wPUM predeployment calibration ICC1:11 3/3 ≥ 0.97 8/8 ≥ 0.98 4/4 ≥ 0.99 6/6 ≥ 0.99 Natural environment pre/postdeployment calibration ICC1:11 3/3 ≥ 0.94 6/8 ≥ 0.90 7/8 ≥ 0.83 for all but one deployment of one monitor3 1/8 = 0.35 4/4 ≥ 0.96 6/6 ≥ 0.98 for all but one deployment of one monitor4 . wPUM Gen 2 Cigalike . wPUM Gen 2 Vapepen . wPUM Gen 2 Hookah . wPUM Gen 2 cigarette . Validated operating range1 Puff flow rate, mL/s 10 to 95 10 to 95 50 to 400 10 to 95 Puff duration2, s 0.2 to 10 0.2 to 10 0.2 to 10 0.2 to 10 Puff gap2, s 0.2 to 10 0.2 to 10 0.2 to 10 0.2 to 10 Puff volume, mL2 6 to 950 6 to 950 45 to 4,000 6 to 950 Accuracy Puff flow rate, mL/s ± 2 ± 2 ± 30 ± 2 Puff duration, s ± 0.1 ± 0.1 ± 0.1 ± 0.1 Puff gap, s ± 0.1 ± 0.1 ± 0.1 ± 0.1 Puff volume, mL ± 2.3 ± 2.3 ± 33 ± 2.3 Resolution Puff flow rate, mL/s 0.1 to 5.0 0.1 to 5.0 0.4 to 20.0 0.1 to 5.0 Puff duration, s 0.025 0.025 0.025 0.025 Puff gap, s 0.025 0.025 0.025 0.025 Puff volume, mL 0.02 to 1.00 0.02 to 1.00 0.08 to 4.00 0.02 to 1.00 Repeatability Number of monitors and deployments 6 deployments across 3 monitors 12 deployments across 8 monitors 16 deployments across 4 monitors 18 deployments across 6 monitors In-Lab PES/wPUM predeployment calibration ICC1:11 3/3 ≥ 0.97 8/8 ≥ 0.98 4/4 ≥ 0.99 6/6 ≥ 0.99 Natural environment pre/postdeployment calibration ICC1:11 3/3 ≥ 0.94 6/8 ≥ 0.90 7/8 ≥ 0.83 for all but one deployment of one monitor3 1/8 = 0.35 4/4 ≥ 0.96 6/6 ≥ 0.98 for all but one deployment of one monitor4 1. There is no upper limit to this parameter. 2For all deployments of each monitor unless noted. 3One monitor had ICC1:1 = 0.18 for one deployment. 4One monitor had ICC1:1 = 0.63 for one deployment. Open in new tab Figure 1. Open in new tabDownload slide Absolute flow rate error. Difference between PES command flow rate and wPUM observed flow rate. Shown are data for the cigalike, vapepen, and cigarette monitors. Figure S2 for the hookah monitor data. Figure 1. Open in new tabDownload slide Absolute flow rate error. Difference between PES command flow rate and wPUM observed flow rate. Shown are data for the cigalike, vapepen, and cigarette monitors. Figure S2 for the hookah monitor data. Figure 2. Open in new tabDownload slide Absolute duration error. Difference between PES command duration and wPUM observed duration. Figure 2. Open in new tabDownload slide Absolute duration error. Difference between PES command duration and wPUM observed duration. Figure 3. Open in new tabDownload slide Absolute Interpuff Gap Error. Difference between PES command interpuff gap and wPUM observed interpuff gap. Figure 3. Open in new tabDownload slide Absolute Interpuff Gap Error. Difference between PES command interpuff gap and wPUM observed interpuff gap. Resolution The flow rate, duration, and volume resolutions are shown in Table 1. The sensitivity of the topography monitors to changes in flow rate increased with increasing flow rate based on the power law theory used to estimate resolution and was different for each monitor due to differences in k. Duration and interpuff gap resolutions were fixed at 0.025 across all monitor types based on the monitor’s sampling frequency (40 Hz). Volume resolutions were computed based on the highest and lowest flow rate resolution and the fixed duration resolution. Repeatability Pairs of pre- and postcalibration coefficients were sampled for about 50% of deployments across the four observation studies (21 monitors across 52 participants). The monitor fleet size for each study varied based on funding available; three cigalike monitors, eight vapepen monitors, four hookah monitors, and six cigarette monitors. The average change in k was 0.5% (ranging from 0.4% to 6.6%) for cigalike monitor deployments, 4.7% (ranging from 0.4% to 75.0%) for the vapepen monitor deployments, 0.1% (ranging from 0.1 to 4.1%) for the hookah monitor deployments, and 1.5% (ranging from 0.1 to 26.3%) for the cigarette monitor deployments. The ICC1:1 was calculated for each pre/postdeployment calibration coefficient as a measure of monitor stability in the natural environment, and is reported in Table 1 for each monitor type. The ICC1:1 was >0.9 for all deployments for 18 of the 21 monitors. Three monitors (two vapepen and one cigarette) had a single deployment with poor repeatability (ICC1:1 of 0.18, 0.35, and 0.63, respectively). In addition, the ICC1:1 was computed for each monitor’s response to the PES command profile at the time of precalibration just prior to deployment, as a measure of variability across monitors in the fleet. The predeployment ICC1:1 for the monitor response compared with the PES command profile was ≥0.97 for all 21 monitors in the fleet. Discussion This study demonstrated the wPUM monitors are stable throughout 1 week natural environment deployment, addressing a current gap in the literature. Specifically, commercially available monitors are unreliable, place restrictions that prevent monitoring over an extended period of time,5 are not compatible with emerging products. The wPUM monitors are compatible with cigarettes, cigalikes, vapepens, including SREC and hookah. The monitors described herein are not compatible with pod style ecigs or super slim cigarettes. This study represents the first critical step toward using topography monitoring to better understand inhaled tobacco products as they are naturally used. Notable considerations, limitations, and future directions are discussed below. Limitations in the Methods Some validation results were limited by the ability of the PES to produce the test profile. The sampling frequency of the PES (20 Hz) was responsible for a minimum error of 0.075 s in duration and interpuff gap. We expect the wPUM cigalike, vapepen, and cigarette monitors can measure volumes on the order of 1 mL and flow rates as low as 2.8 mL/s based on data from prior observational studies,17,18 but accuracy could not be verified because of the lower flow rate limit imposed by the PES. The reported accuracy for all monitors is expected to be more than sufficient for emissions and yield estimates. Considerations in Natural Environment Topography Monitoring Natural environment (NE) monitoring poses challenges not encountered in the lab environment; however, mitigation strategies are available. To address the uncertainly around how the monitor performs while out of sight, monitor calibrations should be checked before and after deployment. Monitors that do not pass pre/postdeployment calibration checks should trigger a full data integrity analysis. In some cases, it is clear from the raw profiles when the monitors become contaminated so that a subset of data can be salvaged.17,18 The risk of fouling increases with the length of deployment and damage may accumulate after multiple deployments. It is important to track the calibration of each monitor over time and implement a maintenance schedule to mitigate flow path contamination. Heavy aerosol producers may require the monitor to be cleaned mid deployment. Alternatively, participants can be given more than one monitor to avoid mid-week maintenance. Monitors may need to be retired after multiple deployments due to wear and odor. Sometimes monitors are damaged, lost or not returned on time. Participant study logs help to piece together questionable topography data and incentives help ensure monitors are returned on time. Another important consideration in NE monitoring is whether participants consistently use the monitor and whether the participant’s topography or use behavior changes because of the monitor. Daily logs are one way to identify and account for noncompliance and exit interviews can assess the users’ impression of how the monitors impacted behavior.17,18 However, objective measures of compliance and altered behavior such as exposure biomarkers are preferred as a means to corroborate subjective measures and verify the monitors captured an authentic account of product use. Finally, special consideration should be taken when attempting to measure topography for cigarettes with filter vent holes. Topography monitors may interfere with compensatory behaviors such as filter vent blocking in light cigarettes, by either covering the vent holes or by making it difficult or awkward for the user to block the vent holes themselves. This scenario highlights the importance of validating the topography monitors with actual observed puffing behavior. Relationship Between Topography and Exposure Puff topography monitors in general (not limited to the monitors tested here) only detect airflow through the tobacco product. They do not detect or analyze the composition of the puff. Emissions depend on the aerosolization effectiveness of the different products and is a function of both puff flow rate and duration. Combined, topography and topography-driven machine emissions provide insight into how users respond to unique product characteristics and how their response impacts exposure. Topography studies show users may change flow rate or other topography parameters3,8,9,26 for different eliquid flavors18 or nicotine levels,27,28 while machine emissions studies show changing flow rate affects the aerosol concentration.19,22 Users taking long puffs with small interpuff gaps may be an attempt to overcome maximum puff durations imposed by ecig manufacturers. Machine emissions from topography playbacks (ie, replay the entire users puffing profile7,29 are needed to better understand these subtle impacts of topography on exposure. Another aspect of exposure not directly measured by topography monitors is respiratory inhalation. Traditional smoking and cigalikes tend to be mouth-to-lung products, meaning users pull a bolus of aerosol into their mouth and then inhale the bolus into their lungs along with clean air. On the other hand, hookah and sub-ohm ecigs tend to be direct-to-lung products, meaning users puff the aerosol directly into their lungs. Mouth-to-lung aerosols are more diluted while direct-to-lung aerosols are highly concentrated. The inhalation maneuver (ie, inhalation and exhalation time and speed, and breath-hold) determines particle deposition in the lung by defining the particle speed and residence time in each airway and how deep in the lung the aerosol travels.30,31,32 Combined, puff topography, inhalation topography, and lung deposition models lead to a better understanding of exposure. Impact on Public Health Policy There are a great number of unanswered questions related to exposure that natural environment topography monitoring can help address. For example, what types of ecig puffing behaviors are most effective at a smoking reduction? What types of ecig topographies are associated with abuse liability? What happens to topography when switching between products, flavors, or nicotine levels? How do changes in behavior affect exposure to nicotine and other constituents? What is the total exposure from the dual use of cigarettes and hookah or dual use of cigarettes and ecigs? If the challenges in NE monitoring can be overcome, the time course of exposure can be tracked as users go about their daily lives, providing data needed to better understand the relative risk of using different inhaled tobacco products. Conclusions This study showed that the wPUM monitors are stable and reliable when deployed in the natural environment, and compatible across a range of tobacco products. Although topography is not a direct measure of exposure, topography does impact emissions and provides an avenue for compensatory behavior and ultimately influences exposure. Natural environment topography monitoring has the potential to address significant gaps in our understanding of the relationship between product characteristics, use behavior, and actual exposure to inform science-based regulatory policy. Supplementary Material A Contributorship Form detailing each author’s specific involvement with this content, as well as any supplementary data, are available online at https://academic.oup.com/ntr. Figure S1. Images of the wPUM family of monitors. Figure S2. Absolute flow rate error for Hookah. Difference between PES command flow rate and wPUM observed flow rate. Declaration of Interests None declared. Acknowledgments The authors acknowledge Ms. Karina Roundtree and Mr. Patrick Morabito for their contributions toward the early development of the wPUM topography monitor family. References 1. Ossip-Klein DJ , Martin JE, Lomax BD, Prue DM, Davis CJ. Assessment of smoking topography generalization across laboratory, clinical, and naturalistic settings . Addict Behav. 1983 ; 8 ( 1 ): 11 – 17 . Google Scholar Crossref Search ADS PubMed WorldCat 2. Mucha RF , Mutz G, Stephan E, Pauli P. Smoking produces a smaller increase in heart rate in the natural smoking environment than in the laboratory . Drug Alcohol Depend. 1996 ; 42 ( 3 ): 209 – 215 . Google Scholar Crossref Search ADS PubMed WorldCat 3. Norton KJ , June KM, O’Connor RJ. Initial puffing behaviors and subjective responses differ between an electronic nicotine delivery system and traditional cigarettes . Tob Induc Dis. 2014 ; 12 ( 1 ): 17 . Google Scholar Crossref Search ADS PubMed WorldCat 4. Behar RZ , Hua M, Talbot P. Puffing topography and nicotine intake of electronic cigarette users . PLoS One. 2015 ; 10 ( 2 ): e0117222 . Google Scholar Crossref Search ADS PubMed WorldCat 5. Mikheev VB , Buehler SS, Brinkman MC, et al. The application of commercially available mobile cigarette topography devices for e-cigarette vaping behavior measurements . Nicotine Tob Res 2018 ;22: nty190 – nty190 . Google Scholar OpenURL Placeholder Text WorldCat 6. Oldham MJ , Plunkett SE, Fisher MT, et al. Laboratory evaluation of the CReSSmicro™ portable topography device: implications for clinical research . Beitr Tab Int 2014 ; 26 ( 1 ): 19 – 25 . Google Scholar OpenURL Placeholder Text WorldCat 7. Hensel EC , Eddingsaas NC, DiFrancesco AG, Jayasekera S, O’Dea S, Robinson RJ. Framework to estimate total particulate mass and nicotine delivered to e-cig users from natural environment monitoring data . Sci Rep. 2019 ; 9 ( 1 ): 8752 . Google Scholar Crossref Search ADS PubMed WorldCat 8. Farsalinos KE , Romagna G, Tsiapras D, Kyrzopoulos S, Voudris V. Evaluation of electronic cigarette use (vaping) topography and estimation of liquid consumption: implications for research protocol standards definition and for public health authorities’ regulation . Int J Environ Res Public Health. 2013 ; 10 ( 6 ): 2500 – 2514 . Google Scholar Crossref Search ADS PubMed WorldCat 9. Hua M , Yip H, Talbot P. Mining data on usage of electronic nicotine delivery systems (ENDS) from YouTube videos . Tob Control. 2013 ; 22 ( 2 ): 103 – 106 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Lopez-Meyer P , Patil Y, Tiffany T, Sazonov E. Detection of hand-to-mouth gestures using a RF operated proximity sensor for monitoring cigarette smoking . Open Biomed Eng J. 2013 ; 9 : 41 – 49 . Google Scholar Crossref Search ADS PubMed WorldCat 11. Behar RZ , Hua M, Talbot P. Puffing topography and nicotine intake of electronic cigarette users . PLoS One. 2015 ; 10 ( 2 ): e0117222 . Google Scholar Crossref Search ADS PubMed WorldCat 12. Goniewicz ML , Kuma T, Gawron M, Knysak J, Kosmider L. Nicotine levels in electronic cigarettes . Nicotine Tob Res. 2013 ; 15 ( 1 ): 158 – 166 . Google Scholar Crossref Search ADS PubMed WorldCat 13. Spindle TR , Breland AB, Karaoghlanian NV, Shihadeh AL, Eissenberg T. Preliminary results of an examination of electronic cigarette user puff topography: the effect of a mouthpiece-based topography measurement device on plasma nicotine and subjective effects . Nicotine Tob Res. 2015 ; 17 ( 2 ): 142 – 149 . Google Scholar Crossref Search ADS PubMed WorldCat 14. Spindle TR , Hiler MM, Breland AB, Karaoghlanian NV, Shihadeh AL, Eissenberg T. The influence of a mouthpiece-based topography measurement device on electronic cigarette user’s plasma nicotine concentration, heart rate, and subjective effects under directed and ad libitum use conditions . Nicotine Tob Res. 2017 ; 19 ( 4 ): 469 – 476 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 15. Ross KC , Juliano LM. Smoking through a topography device diminishes some of the acute rewarding effects of smoking . Nicotine Tob Res. 2016 ; 18 ( 5 ): 564 – 571 . Google Scholar Crossref Search ADS PubMed WorldCat 16. Robinson RJ , Hensel EC, Morabito PN, Roundtree KA. Electronic cigarette topography in the natural environment . PLoS One. 2015 ; 10 ( 6 ): e0129296 . Google Scholar Crossref Search ADS PubMed WorldCat 17. Robinson RJ , Hensel EC, Roundtree KA, Difrancesco AG, Nonnemaker JM, Lee YO. Week long topography study of young adults using electronic cigarettes in their natural environment . PLoS One. 2016 ; 11 ( 10 ): e0164038 . Google Scholar Crossref Search ADS PubMed WorldCat 18. Robinson RJ , Hensel EC, Al-Olayan AA, Nonnemaker JM, Lee YO. Effect of e-liquid flavor on electronic cigarette topography and consumption behavior in a 2-week natural environment switching study . PLoS One. 2018 ; 13 ( 5 ): e0196640 . Google Scholar Crossref Search ADS PubMed WorldCat 19. Robinson RJ , Eddingsaas NC, DiFrancesco AG, Jayasekera S, Hensel EC Jr. A framework to investigate the impact of topography and product characteristics on electronic cigarette emissions . PLoS One. 2018 ; 13 ( 11 ): e0206341 . Google Scholar Crossref Search ADS PubMed WorldCat 20. Farsalinos KE , Spyrou A, Tsimopoulou K, Stefopoulos C, Romagna G, Voudris V. Nicotine absorption from electronic cigarette use: comparison between first and new-generation devices . Sci Rep. 2014 ; 4 : 4133 . Google Scholar Crossref Search ADS PubMed WorldCat 21. Cunningham A , Slayford S, Vas C, Gee J, Costigan S, Prasad K. Development, validation and application of a device to measure e-cigarette users’ puffing topography . Sci Rep. 2016 ; 6 : 35071 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Hensel EC , Sarles SE, Al-Olayan A, et al. A Proposed Waterpipe Emissions Topography Protocol Reflecting Natural Environment User Behaviour . Int J Environ Res Public Health 2019 ; 17 ( 1 ). Google Scholar OpenURL Placeholder Text WorldCat 23. Carcamo N , Robinson R, Pagano T Analysis of four polyaromatic hydrocarbons (PAHs) in cigarette smoke using solid phase extraction and gas chromatography-mass spectrometry (GC-MS) . 237th American Chemical Society National Meeting, Salt Lake City, UT Salt Lake City, UT 2009 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 24. Hensel EC , Jayasekera S, Robinson RJ. Accounting for effects of system dynamics to improve accuracy of emissions reported in e-cig vaping machines . Inhal Toxicol 2018 ;30: 1 – 11 . Google Scholar OpenURL Placeholder Text WorldCat 25. Eddingsaas NC , Hensel EC, O’Dea S, et al. Effect of user puffing topography on total particulate matter, nicotine, and volatile carbonyl emissions from Narghil waterpipes . Tob Control . 2019 ;29(Suppl 2):s117–122. Google Scholar OpenURL Placeholder Text WorldCat 26. Etter JF , Eissenberg T. Dependence levels in users of electronic cigarettes, nicotine gums and tobacco cigarettes . Drug Alcohol Depend. 2015 ; 147 : 68 – 75 . Google Scholar Crossref Search ADS PubMed WorldCat 27. Dawkins LE , Kimber CF, Doig M, Feyerabend C, Corcoran O. Self-titration by experienced e-cigarette users: blood nicotine delivery and subjective effects . Psychopharmacology (Berl.). 2016 ; 233 ( 15-16 ): 2933 – 2941 . Google Scholar Crossref Search ADS PubMed WorldCat 28. Robinson RJ , Hensel EC. Behavior-based yield for electronic cigarette users of different strength eliquids based on natural environment topography . Inhal Toxicol. 2019;31: 13-14 . Google Scholar OpenURL Placeholder Text WorldCat 29. Shihadeh A , Azar S. A closed-loop control “playback” smoking machine for generating mainstream smoke aerosols . J Aerosol Med. 2006 ; 19 ( 2 ): 137 – 147 . Google Scholar Crossref Search ADS PubMed WorldCat 30. Robinson RJ , Yu CP. Deposition of cigarette smoke particles in the human respiratory tract . Aerosol Sci Technol 2001 ; 34 ( 2 ): 202 – 215 . Google Scholar Crossref Search ADS WorldCat 31. NCRP. NCRP Report No. 125 – deposition, retention and dosimetry of inhaled radioactive substances . Bethesda, MD 1997 Natl Council on Radiation (June 1, 1997). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 32. Anjilvel S , Asgharian B. A multiple-path model of particle deposition in the rat lung . Fundam Appl Toxicol. 1995 ; 28 ( 1 ): 41 – 50 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2020. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Characterization and Validation of the Second-generation wPUM Topography Monitors
JO - Nicotine and Tobacco Research
DO - 10.1093/ntr/ntaa153
DA - 2021-01-22
UR - https://www.deepdyve.com/lp/oxford-university-press/characterization-and-validation-of-the-second-generation-wpum-1Gsixq4VT8
SP - 390
EP - 396
VL - 23
IS - 2
DP - DeepDyve
ER -