Compliance in Controlled E-cigarette Studies

Abstract

Introduction

E-cigarette studies have found that the use of a variety of flavors and customizable devices results in greater use frequency and user satisfaction. However, standardized research e-cigarettes are being developed as closed systems with limited flavor options, potentially limiting user satisfaction. In this study, we explore protocol compliance in an e-cigarette study using a standardized, assigned device with puff time and duration tracking (controlled e-cigarette) and potential limitations that controlled devices and e-liquids can introduce.

Methods

In a crossover study, 49 young adult e-cigarette users were recruited using convenience sampling and assigned a controlled e-cigarette device and flavored or unflavored e-liquids on standardized protocols. E-cigarette use frequency (number of puffs per day, collected from the device) and serum cotinine levels were obtained at each of three study visits over 3 weeks. The correlation of cotinine and e-cigarette use over the preceding week was calculated at each study visit.

Results

Correlation of nicotine intake, as measured by serum cotinine, and puff time, as measured by puffs count and duration from the e-cigarette device, as an indicator of study protocol compliance, substantially declined after the first week of the study and were no longer correlated in the remaining study weeks (R² = 0.53 and p ≤ .01 in week 1, R² < 0.5 and p > .05 for remaining weeks).

Conclusions

There is an emerging need for controlled e-cigarette exposures studies, but low compliance in the use of assigned devices and e-liquids may be a limitation that needs to be mitigated in future studies.

Implications

This study is the first to analyze compliance with instructions to use a standardized e-cigarette device with puff time and duration tracking (controlled e-cigarette) across all subjects and an assigned e-liquid flavor over a 3-week period. We find that protocol compliance, as measured by correlations between e-cigarette use measures and cotinine levels, was only achieved in the first week of the study and declined thereafter. These findings indicate that the assignment of a study device and instruction to only use the study device with assigned e-liquid flavor may not be sufficient to ensure participant compliance with the study protocol. We suggest that additional measures, including behavioral and biological markers, are needed to ensure sole use of the study e-cigarette and e-liquid and to be able to interpret results from controlled e-cigarette studies.

Introduction

Studies in the United States have found that the use of flavored e-cigarettes other than tobacco flavors, compared to unflavored, resulted in greater puff volume¹ and reduction in cravings by e-cigarette users.² In addition, myriad devices with variable settings further allow consumers to customize the experience to their liking.² However, developers of standardized e-cigarettes for use in research studies of the effects of e-cigarettes in a more controlled setting are adopting closed system models available only in tobacco flavors (https://www.drugabuse.gov/funding/supplemental-information-nida-e-cig). There is no requirement to use standardized devices in research studies, but well-characterized, standardized e-cigarettes could help to better characterize exposure in e-cigarette users. This study aims to examine compliance with instructions to use such a standardized device.

Methods

The study employed a crossover design, comparing compliance during use of unflavored and flavored e-liquid in a standardized e-cigarette device using a convenience sample.

Participants

Forty-nine participants, who were 18–26 years of age and were heavy e-cigarette users were recruited for this study, conducted between December 2015 and March 2017 in southern California, using online ads and paper recruitment flyers (Table 1). Participants were eligible for participation if they reported e-cigarette use for at least 6 months, use on 6–7 of the last 7 days, and no history of any combustible product use in the past 30 days. Before each study visit, exhaled CO was measured using a BreathCO carbon monoxide monitor (Vitalograph Inc., Lenexa, KS) and procedures were conducted only if CO values were below 10 ppm, indicating no dual use of combustible products. Participants were not trying to quit nicotine use, had no major medical diagnoses, and were not taking asthma medications. Participants used sweet, creamy, buttery, or fruity e-cigarette flavors, and the majority of participants used nicotine concentrations from 3 to 6 mg/mL, which was typical for the e-cigarette market during the time of the study.

Table 1.

Participant Demographics and E-cigarette Use Data (N = 49 Participants)

	Total	Protocol A	Protocol B
	N (%)	N (%)	N (%)
Gender
Female	7 (14.3)	6 (28.6)	1 (3.6)
Male	42 (85.7)	15 (71.4)	27 (96.4)
Ethnicity
Non-Hispanic	36 (73.5)	16 (76.2)	20 (71.4)
Hispanic	13 (26.5)	5 (23.8)	8 (28.6)
Race
Asian	12 (24.5)	2 (9.5)	10 (35.7)
African American	1 (2.0)	1 (4.8)	—
Caucasian	24 (49.0)	13 (61.9)	11 (39.3)
Mixed race	3 (6.1)	2 (9.5)	1 (3.6)
Other	9 (18.4)	3 (14.3)	6 (21.4)
Age
18	1 (2.0)	1 (4.8)	—
19	2 (4.1)	2 (9.5)	—
20	4 (8.2)	4 (19.0)	—
21	4 (8.2)	1 (4.8)	3 (10.7)
22	11 (22.4)	3 (14.3)	8 (28.6)
23	13 (26.5)	5 (23.8)	8 (28.6)
24	5 (10.2)	1 (4.8)	4 (14.3)
25	4 (8.2)	2 (9.5)	2 (7.1)
26	5 (10.2)	2 (9.5)	3 (10.7)
Participant-preferred nicotine concentration (mg/mL)
3	29 (59.2)	12 (57.1)	17 (60.7)
6	17 (34.7)	7 (33.3)	10 (35.7)
12	1 (2.0)	1 (4.8)	—
18	2 (4.1)	1 (4.8)	1 (3.6)
Most frequently used e-liquid flavors
Fruity	22 (44.9)	7 (33.3)	15 (53.6)
Creamy/Buttery/Custard	10 (20.4)	5 (23.8)	5 (17.9)
Menthol/Mint/Wintergreen	3 (6.1)	1 (4.8)	2 (7.1)
Cinnamon	2 (4.1)	1 (4.8)	1 (3.6)
Coffee	—	—	—
Flavorless	—	—	—
Tobacco	—	—	—
Other	12 (24.5)	7 (33.3)	5 (17.9)
	Overall	Protocol A	Protocol B
	Mean (SD)	Mean (SD)	Mean (SD)
Cotinine concentration (ng/mL)
Baseline	117.2 (107.3)	108.6 (90.5)	123.7 (119.6)
Week 1	105.3 (108.3)	89.5 (79.1)	117.2 (126.1)
Week 2	144.3 (187.4)	163.8 (226.1)	129.6 (154.9)
Week 3	116.2 (145)	99.8 (92.2)	128.5 (175.2)
Puff time in previous 48 h (min)
Week 1	14 (13.4)	16.5 (11)	12.1 (14.9)
Week 2	14.5 (12.9)	19.3 (12.7)	11 (12.1)
Week 3	13.1 (16.5)	12.4 (13.1)	13.5 (18.9)

Procedures

The study involved two protocols (A & B), each of 3-week duration, and a total of four study visits, using study-provided e-cigarettes (Figure 1A). At the first visit, e-cigarette liquid preference, nicotine use level, and demographics (based on United States Census Bureau classification) were assessed. The study device and e-liquid were provided per protocol assignment (A & B), as described below and in Figure 1A. For the duration of the study, participants were instructed to use only the Joyetech eVic Supreme battery and Joyetech Delta 23 atomizer, an electronic cigarette that captures the number and duration of puffs. They were also instructed to utilize only the assigned unflavored or flavored e-liquids, per the study protocol, throughout the duration of the study. Participants used study-provided unflavored e-liquid (first 2 weeks), and their own flavored e-liquid (third week) in protocol A or their own flavored e-liquid (first week) and unflavored e-liquids (two subsequent weeks) in protocol B (Figure 1A). Nicotine levels in the unflavored e-liquid used by each participant were based on their preferred flavored commercial e-liquid nicotine concentration reported at the baseline visit (Table 1).

Figure 1.

Study protocol and measures of compliance. (A) Study Protocol Schematic. Participants followed one of two study protocols (A or B) over a period of four study visits. Participants used study-provided unflavored e-liquid (2 weeks), and their own flavored e-liquid (third week) on a study-provided e-cigarette device (Joyetech eVic Supreme battery and Joyetech Delta 23 atomizer). In Protocol A, participants used study-provided unflavored e-liquid for each of the first 2 weeks of the study period (gray) and their preferred flavored e-liquid in the third week (white) (see panel A). In Protocol B, participants used their preferred flavored e-liquid in the first week (white) and study-provided unflavored e-liquid in the second and third weeks (gray). (B–G) Correlation plots of puff time (the sum of the total duration of each puff on the device over the 48 h preceding the visit [in seconds]) against cotinine/nicotine ratio assessed from serum collected at each visit. We computed a standardized estimate of nicotine intake (cotinine/nicotine ratio) by dividing the cotinine level at each visit by the concentration of nicotine in e-liquid solutions for each participant.

Measurements

Time-stamped data on the number and volume of puffs were recorded and downloaded at each study visit from the devices, and used to compute an overall puff duration (number of puffs * volume of each puff). At each visit, serum was also collected to assess cotinine levels. Cotinine levels were assayed by liquid chromatography tandem mass spectrometry using published methods.³

Analysis

A two-way ANOVA was used to compare cotinine and puff time data between weeks and protocols. We computed an estimate of nicotine intake by dividing cotinine levels at each visit by the concentration of nicotine used by each participant. To assess compliance with study device usage, we evaluated the Pearson correlation between nicotine intake and puff time in the 48 hours prior to each visit (Figures 1B–G). The 48-hour time point for puff duration was selected based on cotinine half-life, as cotinine measurements at each visit were likely most associated with use patterns during the previous 48 hours.

Results

Participants mean ± SD and median age was 22.7 ± 1.9 and 23, respectively. Mean ± SD and median preferred nicotine concentration was 4.8 ± 3.3 and 3 mg/mL, respectively. Cotinine levels during the study weeks (week 1: 105.3 ± 108.3 ng/mL, week 2: 144.3 ± 187.3, week 3: 116 ± 145; mean ± SD) did not differ from baseline (117.2 ± 107.3; mean ± SD), or between study protocols. Puff time in the 48 hours prior to study visits, as collected from the study e-cigarette devices, also did not differ between weeks (week 1: 13.9 ± 13.4 minutes, week 2: 14.5 ± 12.9, week 3: 13.1 ± 16.5; mean ± SD) or protocols. The most frequently used flavors during the flavored e-liquid study weeks included fruity, creamy/buttery/custard, menthol/mint/wintergreen, and cinnamon (Table 1).

While all subjects used the study device, we found that compliance with the study protocol substantially declined after the first week of the study as measured by a correlation of nicotine intake and puff time on the assigned device. Compliance was defined as a significant, positive correlation between puff time and cotinine/nicotine ratio with an R² greater than 0.5, per previous correlations observed in other studies.^4,5 Correlation of nicotine intake with puff time was only evident in week 1 of protocol A for aggregate data (R² = 0.53, p ≤ .01; Figure 1B) No correlations were observed for the remaining study weeks (Figure 1C–G), regardless of the randomization of assigned tobacco-flavored or unflavored e-liquid.

Discussion

This study was designed to understand patterns of e-cigarette use and preference for flavored e-liquids using a standardized device in order to limit confounding factors that have been suggested to influence e-cigarette use in other studies, such as varied devices, device settings, and e-liquid components.^4,6–9 Study staff explained to participants that they were only to use the assigned device and e-liquids. However, there was generally poor correlation between puff time and nicotine intake (Figure 1C–G), suggesting that compliance with exclusive study device usage was low. Participants may have used their own devices or e-liquids in addition to the study device. The reason for use of products other than those provided in the study likely include: dislike of the study device or of unflavored e-liquids, desire to use a variety of flavors, or social pressures, including to use a product similar to peers.^10,11

The correlation of puff time with cotinine observed in one of the 6 study weeks (Figure 1B, R² = 0.5) suggests that under limited conditions compliance was reasonably good. The circumstances that may have led the one observed week of compliance are not clear, but may include enthusiasm for study initiation or potentially increased staff emphasis on the importance of only using the unflavored e-liquid that initiated protocol A, compared with starting with flavored e-liquids (protocol B). Low compliance in all other weeks raises significant concerns about subject compliance in long-term controlled vaping studies using a limited selection of flavors and standardized, closed devices. These concerns are especially relevant as previous observational studies^4,5 have shown good correlations between cotinine levels and puffs reported on e-cigarettes.

This study was designed using a standardized device, but allowed participants to use their preferred level of nicotine throughout the study and preferred e-liquid during their flavored e-liquid week. This design was selected to standardize, but minimize change, in individuals’ e-cigarette use behavior to encourage compliance, and to account for differences in device usage across flavored and unflavored weeks using device puff time tracking; however, compliance with study device usage was still poor. Based on our observations, additional strategies are necessary to ensure that controlled, closed system research e-cigarette devices with limited flavoring options, such as NIDA’s Standardized Research E-Cigarette (SREC), which has tobacco or menthol flavors only, are utilized as indicated and without concomitant use of other devices and flavors. The results also indicate that correlation of usage data, such as puff time, and cotinine measurements are useful to assess compliance in interpreting results. As previous observational studies of e-cigarette users have indicated that cotinine levels should be well correlated with puff time,^4,5 we inferred that absence of strong positive correlations indicates a lack of compliance and thus is a useful metric in future studies.

This study has several limitations. Compliance may be better (or worse) if using different designs of controlled research devices, therefore results from future studies using alternative devices may vary from those described here. An ever-changing landscape of new e-cigarette device designs makes the study of e-cigarette use challenging. Our findings are relevant to the devices on the market in 2015–2017; results using currently available products may differ from those reported here. Similarly, newer devices that utilize salt-based nicotine formulations, which have become more popular since the conclusion of this study, may influence compliance in unknown ways. Sampling for this study was limited to one part of the United States, thus these findings may not be representative of other regions of the United States or of other countries. We did not directly ask our subjects about compliance to the study protocol. Finally, no current gold standard exists for assessing compliance with use for e-cigarette studies, so the metrics we utilized here may have unknown limitations.

Additional steps could be taken in future studies to better evaluate compliance with the study protocol. For example, collection of additional biomarkers, such as tobacco-specific nitrosamines, could be used to evaluate the relative dose of e-cigarette use—vs. cigarette use—to improve and monitor compliance in this type of controlled product study. Additionally, considerations for the number of days (or weeks) the participants are required to use an assigned device or flavor could be optimized to enhance compliance, based on the research hypothesis being investigated. Use patterns could be assessed in multiple ways using time-stamped puff recorders, as described here, or with more traditional vaping diaries and detailed use histories. Further, qualitative questions should include direct inquiry about protocol compliance, as merely asking about compliance may improve adherence to study protocol in subsequent weeks.

In conclusion, there is an emerging need for controlled e-cigarette exposures studies in real-world settings, but low compliance in the use of assigned devices and e-liquids may be a limitation that needs to be addressed in these studies. Additional behavioral and biological biomarker assessments are thus needed to ensure adherence to protocols when e-cigarette use variables are limited in future studies.

Acknowledgments

The authors would like to thank Neal Benowitz, MD (Department of Medicine, University of California San Francisco) for his critical review of the manuscript and assistance with the measurement of serum cotinine.

Funding

This work was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (grant number R21HD084812); the National Heart, Lung, and Blood Institute (grant number P50HL120100); and the National Cancer Institute (grant number P50CA180905).

Declaration of Interests

None declared.

Author notes

Co-last authors.