Public Health Implications of Vaping in the US: The Smoking and Vaping Simulation Model

doi:10.21203/rs.3.rs-36677/v1

Download PDF

Research

Public Health Implications of Vaping in the US: The Smoking and Vaping Simulation Model

https://doi.org/10.21203/rs.3.rs-36677/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 17 Apr, 2021

Read the published version in Population Health Metrics →

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Background

Nicotine vaping products (NVPs) are increasingly popular worldwide. They may provide public health benefits if used as a substitute for smoking, but may create public health harms if used as a gateway to smoking or to discourage smoking cessation. This paper presents the Smoking and Vaping Model (SAVM), which estimates the public health implications of NVPs in the US.

Methods

SAVM adopts a cohort-approach and is available on an Excel platform. We derive public health implications by comparing smoking- and NVP-attributable deaths under a No-NVP and an NVP Scenario. The No-NVP Scenario projects current, former and never smoking rates via smoking initiation and cessation rates, and incorporates excess risks of smoking. The NVP Scenario allows for NVP relative excess risks, switching from cigarette to NVP use, separate NVP and smoking initiation rates, and separate NVP and smoking cessation rates. The model is validated against recent US survey data on smoking and vaping prevalence.

Results

The SAVM projects that under current patterns of US NVP use and substitution, NVP use will translate into 5.8 million premature smoking- and vaping-attributable deaths avoided and 96.4 million life years gained between 2013 and 2100. Sensitivity analysis shows that parameters for NVP relative risks, NVP-related switching and smoking cessation are particularly influential in gauging public health impacts.

Discussion

The SAVM shows potential benefits of e-cigarette use over a wide range of parameters. However, there is a high degree of uncertainty regarding key parameters. Policymakers, researchers and other public health stakeholders can apply the SAVM to estimate the potential public health impact of NVPs in their country or region using their own data sources.

Population Biology

E-cigarettes

ENDS

vaping

smoking

simulation model

public health

Smoking prevalence has markedly declined in the US over the past 50 years (Holford et al. 2014b, U.S. Department of Health and Human Services 2014). During that time, cigarette taxes were substantially increased, and smoke-free air policies, media campaigns, youth access policies and smoking cessation treatments were implemented (U.S. Department of Health and Human Services 2014). These policies have been shown to be effective (Levy et al. 2018b), and can explain much of the reduction in smoking prevalence (Holford et al. 2014b, Levy et al. 2016, U.S. Department of Health and Human Services 2014, Warner 2014, Warner et al. 2014). Despite this progress, the harms from cigarette smoking remain unacceptably high (U.S. Department of Health and Human Services 2014). More than 480,000 Americans die each year due to smoking, and two-thirds of all long-term smokers die prematurely of a smoking-attributable disease (Carter et al. 2015, Jha et al. 2013, Thun, Lopez and Hartge 2013). To drastically reduce the harms of smoking on population health, innovative approaches are needed.

The rising popularity of nicotine vaping products (NVPs), also known as e-cigarettes, has been a source of both promise and controversy (Levy et al. 2017). NVPs, especially later generation models, have been shown to more efficiently deliver nicotine (DeVito and Krishnan-Sarin 2018, Farsalinos et al. 2014, Glasser et al. 2016, Wagener et al. 2016, Yingst et al. 2019) and have sensorimotor effects and throat irritation similar to smoking (Etter 2016, Goldenson et al. 2016, Hajek et al. 2018, Voos et al. 2019), thus potentially serving as a substitute for cigarettes. NVPs are increasingly being used by smokers to quit (Caraballo et al. 2017, Filippidis et al. 2017, Giovenco and Delnevo 2018), and recent studies indicate their effectiveness (Bullen et al. 2013, Hajek et al. 2019, Levy et al. 2018d, Zhu et al. 2017). If used by existing continuing smokers switch entirely to NVPs, the health benefits could be substantial since NVPs deliver only a small fraction of the toxicants delivered by smoking (Goniewicz et al. 2014, Goniewicz et al. 2017, Hecht et al. 2014, Ratajczak et al. 2018, Royal College of Physicians 2016, Shahab et al. 2017a, Shahab et al. 2017b). However, there is also growing concern about increased use of these products by youth (Cullen et al. 2019).

The US Food and Drug Administration has the authority to regulate NVPs (Federal Register 2014) and must now consider “the risks and benefits of the tobacco product to the population as a whole”(U.S. Food and Drug Administration 2020). In their review of NVPs, the agency evaluates the toxicity of the product and its effect on the uptake and cessation of existing tobacco products from a public health perspective. The impact of NVP use (“vaping’) on population health will depend on its use in relation to cigarette use (“smoking”) (Levy et al. 2017). As a reduced risk alternative, vaping would improve public health when used by those who would have otherwise initiated smoking or by those who would not have otherwise quit smoking. Vaping would worsen public health if used by those who would not have otherwise initiated NVPs or smoking (i.e., as a gateway) or if used by those who would have otherwise quit smoking.

While many models have been developed that incorporate NVPs (Bachand, Sulsky and Curtin 2018, Cherng et al. 2016, Cobb et al. 2015, Hill and Camacho 2017, Kalkhoran and Glantz 2015, Poland and Teischinger 2017, Soneji et al. 2018, Warner and Mendez 2019, Weitkunat et al. 2015), these models are often geared towards specific applications, lack the flexibility to be easily applied to other settings, require substantial efforts to understand, and have generally not been validated against real-world data. The purpose of this paper is to present a model that considers the public health implications of NVPs and bridges these gaps.

The Smoking and Vaping Model (SAVM) is validated against US observational data on both smoking and NVP use. Unlike previous models, the SAVM adopts a cohort-based approach to better accommodate the impact of new product availability and new policies across birth cohorts. This user-friendly model is based on an earlier model (Levy et al. 2018a), but provides greater flexibility by incorporating a wider range of initiation, cessation and switching parameters and can easily be adapted to other countries, states, or local areas. This study presents the model and demonstrate its application to the US.

Model Structure

The SAVM adopts a public health framework (Levy et al. 2017) that has been specifically developed to consider the impact of NVP use on smoking and public health. We focus on the regular (rather than experimental) use of NVPs and cigarettes and the transitions between those uses both to simplify the model and because health outcomes generally depend on regular (i.e. stable) use patterns over time (Jha et al. 2013). The public health impact of NVP use among smokers and non-smokers is estimated by comparing two scenarios: a) the No-NVP Scenario which projects future cigarette use and associated mortality outcomes for each birth cohort in the absence of NVPs, and b) the NVP Scenario which incorporates NVP use patterns into each cohort’s cigarette use trajectory.

The SAVM begins with separate cohorts of males and females by individual age. Within each cohort, the population evolves with age. To simplify the model and focus on the current, former and never smoking and vaping, their rates are translated into population levels using the actual and projected US population (Centers for Disease Control and Prevention 2019a).

The No-NVP Scenario

The No-NVP Scenario projects the prevalence of never, current, and former smokers over time using age and gender-specific initiation and cessation rates for each cohort. The data were developed using an age-period-cohort statistical smoking model (Holford et al. 2014a, Holford, Levy and Meza 2016, Jeon et al. 2018, Tam et al. 2018). Established smoking is measured as having smoked at least 100 cigarettes during one’s lifetime and currently smoking every day or some days. Initiation is based on becoming an established current smoker and occurs through age 40. Current smokers become former smokers after having quit. A two year cut-off is applied to reflect cessation net of relapse, i.e., relapse is not explicitly considered. The initiation and cessation rates are projected forward based on data from the NHIS through the year 2013, before NVPs became popular and thus reflects smoking patterns in the absence of NVPs.

For a given cohort, the projected prevalence of never smokers at age a and time t is composed of the surviving never smokers in the previous year (age a-1, time t-1) who did not initiate smoking. Denoting the initiation rate from never smokers at a-1 and t-1 to smokers at a and t as Init_a−1,t−1 and the death rate of never smokers in previous year as DR_never_a−1,t−1, never smoker prevalence at age a and year t is calculated as:

Prev_never_a,t= Prev_never_a−1,t−1 * (1 - DR_never_a−1,t−1) * (1 – Init_a−1,t−1). (1)

The prevalence of smokers includes surviving never smokers who initiate smoking and surviving smokers who do not quit. Denoting the cessation rate from smokers at a-1 and t-1 to former smokers at a and t as Cess_a−1,t−1 and the smoker death rate as DR_smoker_a−1,t−1, current smoking prevalence evolves as:

Prev_smoker_a,t = Prev_smoker_a−1,t−1 * (1- DR_smoker_a−1,t−1) * (1- Cess_a−1,t−1)

+ Prev_never_a−1,t−1 * (1 - DR_never_a−1,t−1) * Init_a−1,t−1. (2)

Denoting DR_former as the former smoker death rate, former smokers include surviving former smokers and surviving smokers who quit smoking for at least two years, former smoking prevalence is:

Prev_former_a,t = Prev_former_a−1,t−1 * (1 – DR_former_a−1,t−1) + Prev_smoker_a−1,t−1

* (1- DR_smoker_a−1,t−1) * Cess_a−1,t−1. (3)

Since prevalence estimates decline each year due to deaths, each of the prevalence rates is multiplied by the following correction factor to assure that they sum to 100%:

Correction factor_a,t = 1/[Prev_never_a−1,t−1*(1-DR_never_a−1,t−1) + Prev_smoker_a−1,t−1*

(1-DR_smoker_a−1,t−1) + Prev_former_a−1,t−1*(1-DR_former_a−1,t−1)]. (4)

The NVP Scenario

Starting from the same initial current, former and never smoking prevalence as in the No-NVP Scenario, the NVP Scenario is expanded to include current and former NVP use. The transitions are shown in Fig. 1.

Figure 1 here

A switching parameter allows direct transitions from smoking to vaping. This parameter varies by age and varies over time through an exponential decay process. The model directly relates the NVP Scenario to the No-NVP Scenario by defining the NVP and smoking initiation rates in the NVP Scenario through separate linear multipliers applied to the smoking initiation rate (Init_a.t) in the No-NVP Scenario. Using a linear multiplier assures that NVP and smoking initiation follows the same age patterns for smoking initiation as in the No-NVP case, e.g., smoking initiation mostly occurs before age 21 and is minimal after age 30. Similarly, NVP and smoking cessation are modelled as separate linear multipliers of smoking cessation (Cess_a,t) in the No-NVP Scenario, so that cessation follows the same age pattern in the No-NVP Scenario and tends to increase with age. With variations by age incorporated via the respective initiation and cessation patterns in the No-NVP Scenario and because current studies do not indicate different patterns, the smoking and vaping initiation and cessation multipliers are simply modeled as independent of age. In addition, these multipliers are assumed to remain constant over time, so that smoking and vaping initiation and cessation in the NVP Scenario follow the same temporal patterns as in the No-NVP Scenario.

Because SAVM focuses on regular use, the NVP Scenario does not explicitly incorporate “gateway” effects whereby youth transition from short-term vaping to regular smoking. Any such transitions are however incorporated indirectly through the smoking initiation multiplier in the NVP Scenario. Similarly, the NVP Scenario does not explicitly model short-term NVP use leading to quitting both cigarettes and NVPs, but is indirectly incorporated through the smoking cessation multiplier. In addition, the NVP Scenario does not distinguish dual cigarette and NVP users from exclusive smokers, because dual user have generally been found to either remain dual users or soon transition to either exclusive cigarette or exclusive NVP use (Azagba, Shan and Latham 2019, Borland et al. 2019, Coleman et al. 2018, Robertson et al. 2019, Stanton et al. 2020, Taylor et al. 2020) and the health risks of dual users are similar to those of exclusive smokers (Goniewicz et al. 2017, Shahab et al. 2017a, Shahab et al. 2017b). However, transitions from dual use to exclusive NVP use or to neither cigarette or NVP use are reflected in the switching and smoking cessation parameters.

Never smokers include surviving never smokers who do not initiate into smoking or NVP use, calculated as:

Prev_never_a,t = Prev_never_a−1,t−1 *(1 – DR_never_a−1,t−1)* (1 – α_S* Init_a−1,t−1– α_V * Init_a−1,t−1), (5)

where α_S and α_V are the smoking (S) and vaping (V) initiation multipliers. Smokers include surviving smokers who do not switch to vaping and who do not quit smoking and surviving never smokers who initiate smoking:

Prev_smoker_a,t = Prev_smoker_a−1,t−1*(1-DR_smoker_a−1,t−1)*(1 – σ_a−1*(1-δ)^t−t0–1 – χ_S* Cess_a−1,t−1)

+ Prev_never_a−1,t−1 * (1–DR_never_a−1,t−1) *(α_S*Init_a−1,t−1), (6)

where σ_a−1 is the switching rate from smoking to NVP use in the initial year t₀ with decay rate δ for each cohort and χ_S is the smoking cessation multiplier. Exclusive NVP users include surviving never smokers who initiate vaping, surviving vapers who do not quit, and those switching to vaping from surviving current smokers before age 35:

Prev_NVP_a,t = Prev_never_a−1,t−1*(1-DR_never_a−1,t−1)*α_V*Init_a−1,t−1 + Prev_NVP_a−1,t−1*

(1-DR_NVP_a−1,t−1)*(1– χ_V * Cess _a−1,t−1) + Prev_smoker_a−1,t−1* I(a-1 < 35)*

(1–DR_smoker_a−1,t−1)* σ_a−1*(1-δ)^t−t0–1_, where I(a-1) = 1 if a < 36, otherwise = 0. (7)

DR_NVP is the death rate of exclusive NVP users and χ_V is the NVP cessation multiplier. The transition from smoker to exclusive NVP user is conditional on being less than age 35, because mortality risks of smoking are small when quitting before age 35 (Jha et al. 2013).

Former smokers include surviving former smokers, surviving smokers who quit smoking and do not vape, and surviving former smokers who regularly vaped but quit vaping:

Prev_former_a,t = Prev_formers_a−1,t−1*(1 – DR_former_a−1,t−1) + Prev_smoker_a−1,t−1 *

(1 – DR_smoker_a−1,t−1) * (χ_S* Cess _a−1,t−1) + Prev_FS-NVP_a−1,t−1 *

(1 – DR_FS-NVP_a−1,t−1) * (χ_V* Cess _a−1,t−1), (8)

where Prev_FS-NVP_a−1,t−1 is the prevalence of former smokers who currently use NVPs and DR_FS-NVP_a−1,t−1 is the corresponding death rate. Former NVP users include surviving former NVP users and surviving vapers who quit:

Prev_FNVP_a,t = Prev_FNVP_a−1,t−1 * (1- DR_FNVP_a−1,t−1) +

Prev_NVP_a−1,t−1 * (1 – DR_NVP_a−1,t−1) * χ_V * Cess_a−1,t−1, (9)

where DR_FNVP_a−1,t−1 denotes the former NVP user death rate. Since former smokers using NVPs may quit vaping, former smokers using NVPs (after age 35) includes surviving former smokers using NVPs who do not quit vaping and smokers who switch to vaping:

Prev_FS-NVP_a,t = Prev_FS-NVP_a−1,t−1*(1- DR_FS-NVP_a−1,t−1)*(1 – χ_V * Cess _a−1,t−1)

+ Prev_smoker_a−1,t−1* I(a-1 > 35) * (1- DR_smoker_a−1,t−1)*(σ_a−1*(1 + δ)^t−t0–1),

where I(a-1) = 1 if a ≥ 36, otherwise = 0 (10)

As in the No-NVP Scenario, a correction factor is applied as the reciprocal of the sum of prevalence rates for each smoking and vaping category i to ensure that the prevalence rates each period sum to 100%:

Correction factor_a,t = 1 / ∑_i [Prev_i_a−1,t−1 * (1 – DR_i_a−1,t−1)]. (11)

Public Health Outcomes

The SAVM considers two public health outcomes considered: 1) Smoking- and vaping-attributable deaths, and 2) smoking- and vaping-attributable life-years lost (LYLs).. Both are based on the excess risks of smoking or vaping and the number of current and former smokers and vapers.

In the No-NVP Scenario, smoking-attributable deaths (SADs) for current smokers is calculated by applying the excess risks of smokers relative to never smokers (DR_smoker_a,t – DR_never_a,t) to the smoking population:

SAD_smoker_a,t = (DR_smoker_a,t – DR_never_a,t) * Prev_smoker_a,t * Population_a,t, (12)

and attributable deaths for former smokers is calculated as:

SAD_former_a,t = (DR_former_a,t – DR_never_a,t) * Prev_former_a,t * Population_a,t (13)

SADs for current and former smokers are summed over all ages in a particular year to obtain total SADs in that year. LYLs are estimated as the number of premature deaths multiplied by the remaining life expectancy (LE) of a never smoker at each age in year t denoted by LE_Never_a,t:

LYL_a,t = (SAD_smoker_a,t + SAD_former_a,t ) * LE_Never_a,t. (14)

Summing over ages in a particular year obtains the LYLs in that year.

In the NVP Scenario, SAD_smoker_a,t and SAD_former_a,t are calculated in the same manner as above except using their respective prevalence rates in the NVP scenario. Vaping-attributable deaths for current exclusive NVP users (VADs) and for former NVP users (FNVP) are measured by two similar equations (15) and (16). For current exclusive NVP users, attributable deaths (VAD_NVP_a,t) are a product of the number of NVP users and the excess risks of smoking adjusted by NVP relative risk, denoted by Risk_NVP, and similarly for former NVP use attributable deaths (VAD_FNVP_a,t):

VAD_NVP_a,t = Risk_NVP*(DR_smoker_a,t - DR_never_a,t)*Prev_NVP_a,t*Population_a,t. (15)

VAD_FNVP_a,t = Risk_NVP*(DR_former_a,t -DR_never_a,t)*Prev_FNVP_a,t*Population_{a,t. (}16)

As a special case in the NVP Scenario, the death rate of NVP users who previously smoked is determined by the portion of excess risk of current relative to former smokers plus the death rate of former smokers:

DR_FS-NVP_a,t = Risk_NVP * (DR_smoker_a,t - DR_former_a,t) + DR_former_a,t (17)

Smoking and vaping-attributable deaths (SVADs) for former smokers who currently use NVPs (FS-NVPs) in the NVP Scenario is measured as a product of the number of FS-NVPs and the excess mortality risk of FS-NVP to never smokers:

SVAD_FS-NVP_a,t = [DR_FS-NVP_a,t -DR_never_a,t)]*Prev_FS-NVP_a,t* Population_a,t (18)

Attributable deaths are calculated by summing SADs, VADs and SVADs over ages each year for a given cohort. LYLs in the NVP Scenario at each age are calculated as:

LYL_a,t = (SAD_smoker_a,t + SAD_former_a,t + VAD_NVP_a,t + VAD_FNVP_a,t +

SVAD_FS-NVP_a,t)*LE_Never_a,t (19)

Summing over ages in a particular year obtains the LYLs in that year.

The public health impact of NVP use each year is evaluated as the difference in attributable deaths between the No-NVP and NVP Scenarios, and similarly for LYLs.

Data and Parameters

A detailed description of model parameters is shown in Table 1.

Table 1

Data and Initial Parameters for the US Smoking and Vaping Model
Input parameters	Description	Data source or estimate
Population	Population by age, gender, and year (2013–2100)	US population projections (Center for Disease Control and Prevention 2019)
Mortality rates	Mortality rates by age, gender, and year for never, current, and former smokers (2013–2100)	CISNET Lung Group,(Holford et al. 2014a, Holford, Levy and Meza 2016, Jeon et al. 2018, Tam et al. 2018) available on the CISNET website (CISNET 2016).
Expected life years	Expected life years remaining of never smokers by age, gender, and year (2013–2100)	CISNET Lung Group,(Holford et al. 2014a, Holford, Levy and Meza 2016, Jeon et al. 2018, Tam et al. 2018) available on the CISNET website (CISNET 2016).
Smoking prevalence	Current and former smoking prevalence by age and gender for initial year	CISNET Lung Group,(Holford et al. 2014a, Holford, Levy and Meza 2016, Jeon et al. 2018, Tam et al. 2018) available on the CISNET website (CISNET 2016).
NVP relative risk multiplier	Excess risk of NVP use measured relative to excess smoking risks (mortality rate of current smokers – mortality rate of never smokers)	NVP mortality risks estimated to be 5% that of smoking excess risk for both genders at all ages, based on multi-criteria decision analysis(Nutt et al. 2016) and independent review (McNeill et al. March 2020). 40% risk is also applied in the sensitivity analysis.
NVP Switching rate	Rate of switching from smoking cigarettes to exclusive NVP use	Ranges from 0.6%-4.0%, estimated by age group and gender using prospective analysis from PATH data 2013–2018
Smoking initiation multiplier in the NVP Scenario	Ratio of smoking initiation rate in the NVP Scenario to smoking initiation in the No-NVP Scenario	75% of No-NVP smoking initiation rate, based on recent studies (Levy, Yuan and Li 2017, Levy et al. 2018c, National Center for Health Statistics 2019).
NVP initiation multiplier in the NVP Scenario	Ratio of NVP initiation rate in the NVP Scenario to smoking initiation in the No-NVP Scenario	50% of No-NVP smoking initiation rate, based on recent studies (Cullen et al. 2019, Hammond et al. 2019, Miech et al. 2019b, Miech et al. 2019c).
Smoking cessation multiplier in the NVP Scenario	Ratio of smoking cessation rate in the NVP Scenario to smoking cessation in the No-NVP Scenario	100% of the No-NVP smoking cessation rate
NVP cessation multiplier in the NVP Scenario	Ratio of NVP cessation rate in the NVP Scenario to smoking cessation in the No-NVP Scenario	100% of the No-NVP smoking cessation rate
Notes: NVP = nicotine vaping product. No-NVP Scenario refers to values in the absence of NVP use. NVP Scenario refers to values with NVP use.

Table 1 here

The No-NVP Scenario

The current, former and never smoking death rates and expected life years of never smokers were developed by Rosenberg et al (Holford et al. 2014a, Holford et al. 2014b, Holford, Levy and Meza 2016, Rosenberg et al. 2012) and are available on the CISNET website (CISNET 2016). Current population, mortality rate and expected life year estimates are from the CDC (Center for Disease Control and Prevention 2019).

The NVP Scenario

The NVP Scenario requires six input parameters: NVP mortality risks, NVP switching rate, smoking initiation, vaping initiation, smoking cessation and vaping cessation.

The NVP relative risk multiplier, Risk_NVP, represents the relative risk of death associated with current NVP use as a percentage of the excess mortality risk experienced by current or former smokers as defined above. NVP relative mortality risks are designated as 5% of cigarette excess risks based on a multi-criteria decision analysis (Nutt et al. 2016) and a Public Health England review (McNeill et al. March 2020). Since others have suggested higher risks (de Groot et al. 2018, Nelluri et al. 2016, Tegin et al. 2018, Unger and Unger 2018), an NVP relative risk multiplier of 40% is also considered.

The NVP switching rate σ_a is the annual rate at which current smokers switch from smoking to NVP use, leading to a direct reduction in smoking prevalence. Baseline male (female) NVP yearly switching rates are based on prospective data that we collected from the 2013/2014, 2014/2015, 2015/2016, and 2016/2017 Population Assessment of Tobacco and Health (PATH) surveys (U.S. Food and Drug Administration, National Institutes of Health and Abuse. 2019) averaged over years: 4% (2.5%) for ages 18–24, 2.5% (2.0%) for ages 25–34, 2.5% (1.6%) for age 35–44, 1.3% (1.4%) for ages 45–54, 1.2% (1.4%) for ages 55–64, and 0.6% (1.0%) for ages 65 and above. Those below age 18 were assigned the age 18–24 rate. We also consider switching rates that are 50% lower and 100% higher than those baseline estimates. Rates are initially assumed constant over time (i.e., 0% decay), but we also consider a 20% decay rate, δ, to reflect the potential for switching rates to decline in the absence of innovation.

Since the smoking initiation in the NVP Scenario is measured relative to smoking initiation (Init_a,t) in the No-NVP Scenario, the smoking initiation multiplier α_S is greater than 100% if NVP use increases net smoking initiation beyond the rate at which individuals would have otherwise initiated smoked in the absence of NVPs (i.e., gateway in > gateway out), and is less than 100% if those who would have initiated smoking tend to transition to exclusive NVP use instead of smoking (i.e., gateway out > gateway in). Based on the more rapid downward trend in US youth and young adult smoking as NVP use increased in recent years (Levy et al. 2019a), the age- and time-invariant baseline smoking initiation multiplier is initially set at 75%, i.e. a 25% net decrease in smoking initiation due to vaping. A range of 25–125% is also considered.

The NVP initiation rate multiplier α_V reflects the youth and young adult initiation of NVP use relative to smoking initiation rates (Init_a,t) in the No-NVP Scenario. If less than 100%, NVP initiation rates are lower than smoking initiation rates in the No-NVP Scenario. i.e., fewer individuals become regular NVP users than who would have become smokers in the No-NVP Scenario. Reflecting the increased US youth and young adults regular NVP users in recent years (Cullen et al. 2019, Hammond et al. 2019, Levy, Yuan and Li 2017, Miech et al. 2019b, Miech et al. 2019c), the age- and time-invariant baseline NVP initiation multiplier is initially set at 50% for all ages. Sensitivity analysis is conducted over the range of 25–75%.

Since smoking cessation in the NVP Scenario is measured relative to smoking cessation (Cess_a,t) in the No-NVP Scenario, the smoking cessation multiplier χ_S is greater than 100% if smoking cessation rates are higher in the NVP Scenario than in the No-NVP Scenario, e.g., if the availability of NVPs leads to increased smoking cessation. A parameter less than 100% implies that smokers on net are less likely to quit smoking in the NVP Scenario compared to the No-NVP Scenario, e.g., NVP use leads to dual use (continued smoking) rather than complete cessation among smokers. The baseline smoking cessation rate multiplier is initially set at 100% with a range of 50%-150% also considered.

The NVP cessation multiplier χ_V is greater than 1 if NVP cessation rates are higher than smoking cessation rates in the No-NVP Scenario. The parameter is less than 100% if NVP cessation rates are lower than smoking cessation rates in the No-NVP Scenario, e.g., NVPs are less addictive than cigarettes. The baseline NVP cessation multiplier is set at 100%, with a plausible range of 50%-150%.

Model Validation and Analysis

The model estimates the NVP effects over time for the prevalence of current and former smokers, current and former NVP users, and former smokers using NVPs, and for smoking-attributable and NVP-attributable deaths. The model was first validated over the years 2013 to 2018 by comparing model predictions of current smoking prevalence to current smoking prevalence rates from the NHIS (Centers for Disease Control and Prevention 2019b) by age and gender. We focus on relative reductions over the period 2013–2018, because levels in the initial year 2013 in the model differ from those in the NHIS. We also validated NVP use. However, we confine the model comparison to NVP prevalence measures from the 2018 NHIS. Although NVP use was already occurring in 2013, the model itself begins with no NVP use that year and is adjusted to reach the 2018 levels. (Levy et al. 2019b). Consequently, we validate NVP use against NHIS estimates for 2018, the latest year for which data was available rather than rely on prior trend data. We defined NVP users as those who used e-cigarettes at least 10 of the past 30 days to reflect more regular use.

Upon validating the model, we consider how NVP use affects smoking prevalence, smoking- and NVP-attributable deaths and LYLs. We conduct sensitivity analyses for the NVP transition and risk parameters over the plausible ranges specified above focusing on their impact on premature attributable deaths and LYLs.

Validation of Smoking Prevalence

Using the initial (best estimate) input parameter estimates, the SAVM predictions are validated against smoking prevalence from NHIS data for 2013–2018 as shown in Table 2. For ages 18 and above, SAVM projects that male (female) smoking prevalence fell from 21.4% (15.9%) in 2013 to 16.7% (12.6%) in 2018, while NHIS shows a decline from 20.3% (15.4%) in 2013 to 15.8% (12.0%) in 2018. NHIS shows a 22.2% (22.1%) relative reduction between 2013 and 2018 compared to a 22.2% (20.7%) relative reduction in SAVM.

Table 2 here

By age group, SAVM projects that male (female) prevalence for ages 18–24 fell by 33.0% (33.7%) in relative terms between 2013 and 2018 compared to 60.8% (52.6%) from NHIS over the same time period. For ages 25–44, SAVM projects a relative decline of 21.0% (17.6%) compared to 17.2% (16.6%) from NHIS. For ages 45–64, SAVM projects a relative decline of 19.9% (18.1%) compared to 15.2% (20.9%) from NHIS. For ages 65 and above, SAVM projects a relative decline of 16.6% (20.3%) compared to 6.3% (4.0%) from NHIS.

Thus, while predicting smoking prevalence well for the adult population, SAVM tends to under-predict relative reductions in smoking prevalence at younger ages and over-predict relative reduction at older ages. While SAVM estimates for 2018 are often outside the NHIS confidence intervals, upon scaling all years by the initial year (2013) from SAVM relative to the initial year in NHIS for comparability, the SAVM 2018 estimates are generally within the NHIS confidence intervals, with the exceptions of the 18–24 age group for males and the 65 and above age group for females. The scaled age group for ages 18 and above is shown for in Figs. 2.a. (includes the SAVM and NHIS (with 95% confidence interval) male prevalence rates as shown in Table 2, and the scaled male SAVM prevalence rates calculated as the male SAVM prevalence rates in 2013–2018 multiplied by 94.8% =2013 NHIS male rate/2013 SAVM male rate) and Fig. 2.b. (includes the SAVM and NHIS (with 95% confidence interval) female prevalence rates as shown in Table 2, and the scaled female SAVM prevalence rates calculated as the as the female SAVM prevalence rates in 2013–2018 multiplied of 96.7% =2013 NHIS female rate/2013 SAVM female rate).

Figure 2.a. and Fig. 2.b. here

Validation of NVP Prevalence

The SAVM predictions for exclusive NVP use are validated against the overall NVP prevalence from NHIS data for 2018 as shown in Table 3. For ages 18 and above, SAVM projects 2018 NVP prevalence of 3.0% for males compared to 3.1% (95% CI:2.7%-3.5%) from NHIS, and 1.8% for females compared to 1.5% (95% CI:1.3%-1.8%). By age group for males, the SAVM projection against NHIS comparisons (with confidence intervals) for 2018 are 7.5% vs 6.8% (95% CI:4.8%-8.8%) for ages 18–24, 4.2% compared to 4.7% (95% CI:3.8%-5.6%) for ages 25–44, 1.4% vs. 1.5% (95% CI:1.0%-1.9%) for ages 45–64, and 0.5% vs. 0.6% (95% CI:0.3%-0.9%) for ages 65 and above. For females, SAVM projections vs NHIS estimates are 4.9% vs. 3.1% (95% CI:1.8%-4.5%) for ages 18–24, 2.4% vs. 1.9% (95% CI:1.4%-2.3%) for ages 25–44, 1.0% vs. 1.3% (95% CI:1.0%-1.7%) for ages 45–64, and 0.4% vs. 0.5% (95% CI:0.3%-0.7%) for ages 65 and above. SAVM predictions (overall and by age group) fell within the 95% confidence intervals estimated from NHIS data, except for females ages 18–24 and 25–44.

Table 3. here

Public Health Impact Under the Baseline NVP and No-NVP Scenario

Table 4 presents US smoking and NVP use prevalence, deaths, and LYLs for males and females ages 18–99 over the modeling period 2013–2100 for all existing and new birth cohorts. We first apply the baseline parameters: Risk_NVP =5% that of smoking, with multipliers for smoking initiation α_S = 75%, NVP initiation α_V = 50%, smoking cessation χ_S = 100%, NVP cessation χ_V = 100%, with switching rates σ_a remaining constant over time (δ = 0%).

Table 4

Baseline SAVM outcomes under NVP Scenario vs. No-NVP Scenario, all cohorts including new births, US, by gender, in years 2013–2100, ages 18–99
	Year	2013	2014	2040	2100	Cumulative^a
US Males
No-NVP Scenario^b	Smokers (%)	21.4%	20.9%	13.7%	11.5%	-
	SADs	280,925	279,542	266,482	157,507	19,945,214
	LYLs	3,668,313	3,651,554	2,808,307	1,665,429	217,668,305
NVP Scenario^c	Smokers (%)	21.4%	20.4%	6.3%	4.0%	-
	NVP users (%)	0.0%	0.4%	7.2%	8.9%	-
	FS-NVP users (%)	0.0%	0.2%	2.2%	1.2%	-
	SVADs	280,925	278,242	234,175	65,057	15,591,862
	LYLs	3,668,313	3,623,321	2,080,017	585,121	144,047,497
Difference	SADs averted	0	1,300	32,307	92,450	4,353,351
Difference	LYLs averted	0	28,233	728,290	1,080,309	73,620,808
US Females
No-NVP Scenario	Smokers (%)	15.9%	15.5%	9.8%	8.2%	-
	SADs	110,815	109,395	109,013	57,578	7,803,840
	LYLs	1,405,272	1,399,722	1,079,004	569,181	80,719,695
NVP Scenario	Smokers (%)	15.9%	15.2%	5.5%	3.7%	-
	NVP users (%)	0.0%	0.2%	4.4%	5.5%	-
	FS-NVP users (%)	0.0%	0.1%	1.4%	0.8%	-
	SVADs	110,815	108,576	95,375	27,797	6,266,913
	LYLs	1,405,272	1,385,563	829,318	233,051	56,528,536
Difference	SADs averted	0	819	13,638	29,782	1,536,927
Difference	LYLs averted	0	14,159	249,686	336,130	24,191,158
Both genders
Difference	SADs averted	0	2,119	45,946	122,232	5,890,278
	LYLs averted	0	42,392	977,976	1,416,439	97,811,967
	SADs averted (%)	0.0%	0.5%	12.2%	56.8%	21.2%
	LYLs averted (%)	0.0%	0.8%	25.2%	63.4%	32.8%
Notes: NVP = nicotine vaping product, SADs = smoking attributable deaths, SVADs = smoking and vaping attributable deaths, LYLs = life-years lost, FS-NVP = former smokers using NVPs.
^a Cumulative is the sum of SADs/SVADs or LYLs over the years 2013–2100.
^b No-NVP Scenario refers to the values of smoking prevalence (%), SADs, and LYLs in the absence of NVP use.
^c NVP Scenario refers to values of smoking prevalence (%), exclusive NVP prevalence (%), former smokers using NVP prevalence (%), SVADs, and LYLs with NVP use.

Table 4 here

In the No-NVP Scenario, smoking prevalence ages 18 and above is 21.4% in 2013 declining to 13.7% in 2040, and 11.5% in 2100 for males, 15.9% in 2013 declining to 9.8% in 2040, and 8.2% in 2100 for females. Cumulative (2013–2100) SADs are 19.9 million for males and 7.8 million for females, while cumulative LYLs are 217.7 million for males and 80.7 million for females.

In the NVP Scenario, male (female) smoking prevalence declines to 6.3% (5.5%) in 2040 and then declines to 4.0% (3.7%) in 2100. Exclusive NVP users include those who initiated NVP use from never smokers (de novo) at all ages and those who switched to NVP use from smoking before age 35. Male (female) exclusive NVP use increases from 0.4% (0.2%) in 2014 to 7.2% (4.4%) in 2040, and to 8.9% (5.5%) in 2100. The prevalence of former smokers using NVPs (those switching from smoking to vaping after age 34) increases from 0.2% (0.1%) in 2014 to 2.2% (1.4%) in 2040, then declines to 1.2% (0.8%) in 2100. Compared with females, greater reduction in male smoking and increase in NVP use is due to higher levels of switching to NVPs at younger ages among males. Smoking- and vaping-attributable deaths (SVADs) and LYLs cumulate to 15.6 million and 144.0 million during 2013–2100 for males and to 6.3 million and 56.5 million for females.

Under the NVP Scenario, approximately 5.9 million (4.4 million males, 1.5 million females) premature deaths are averted, a 21% relative reduction compared to the No-NVP Scenario, and 97.8 million (73.6 million males, 24.2 million females) LYLs are averted, a 33% relative reduction.

Sensitivity to NVP Transition Parameters

Sensitivity analyses are shown in Table 5 for smoking- and vaping-attributable deaths (SVADs) and in Table 6 for life-years lost (LYLs). While holding the NVP relative risk multiplier (Risk_NVP) constant at 5% (columns 2 and 3 in both tables), we examine the individual impact of variations in transition rates on SVADs and LYLs averted relative to the initial levels in the NVP Scenario. Each of five transition multipliers were varied for 1) NVP switching, 2) smoking initiation, 3) NVP initiation, 4) smoking cessation, and 5) NVP cessation.

Table 5

Sensitivity analysis: Smoking attributable deaths and deaths averted in the No-NVP Scenario and NVP Scenario across parameter changes, both genders, ages 18–99, 2013–2100
Scenario	NVP relative risk (Risk_NVP) ^a = 5%		NVP relative risk (Risk_NVP) = 40%		Relative Change (5% vs 40%)^d
No-NVP Scenario	Total SADs		Total SADs
No-NVP Scenario	27,749,053	-	27,749,053	-
NVP Scenario with parameter changes from baseline	Averted SADs and SVADs^b	Relative Change (vs. baseline estimate)^c	Averted SADs and SVADs^b	Relative Change (vs. baseline estimate)^c
50% of switch rate (σ_a),^f no decay (δ = 0%) ^g	5,890,278	-	2,947,060	-	-50.0%
200% of switch rate (σ_a), no decay (δ=0%)	4,195,978	-28.8%	1,946,856	-33.9%	-53.6%
100% of switch rate (σ_a), decay (δ= −0.20%)	7,914,890	34.4%	4,141,413	40.5%	-47.7%
25% of smoking initiation multiplier (α_S)^h	2,332,235	-60.4%	835,516	-71.6%	-64.2%
125% of smoking initiation multiplier (α_S)	7,351,025	24.8%	4,913,944	66.7%	-33.2%
50% of switch rate (σ_a),^f no decay (δ = 0%) ^g	4,588,952	-22.1%	1,186,597	-59.7%	-74.1%
200% of switch rate (σ_a), no decay (δ=0%)	5,901,511	0.2%	3,483,735	18.2%	-41.0%
100% of switch rate (σ_a), decay (δ= −0.20%)	5,880,955	-0.2%	2,441,568	-17.2%	-58.5%
50% of smoking cessation (χ_S)^j	3,615,711	-38.6%	289,152	-90.2%	-92.0%
150% of smoking cessation (χ_S)	7,081,194	20.2%	4,395,398	49.1%	-37.9%
50% of NVP cessation (χ_V)^k	5,636,131	-4.3%	1,209,331	-59.0%	-78.5%
150% of NVP cessation (χ_V)	5,998,902	1.8%	3,717,187	26.1%	-38.0%
Notes: NVP = nicotine vaping product, SADs = smoking attributable deaths.
^a The NVP relative risk multiplier is the mortality risk of NVPs as a percentage of the excess mortality risk of smoking.
^b The absolute reduction in smoking- and vaping-attributable deaths in the NVP Scenario compared to the No-NVP Scenario over 2013–2100.
^c The relative percent change in averted SADs for each NVP Scenario compared to the initial NVP Scenario (best estimate). A negative (positive) value means that changing the parameter will decrease (increase) the averted SADs in the specific scenario relative to averted SADs in the initial NVP Scenario.
^d The relative percent change in averted SADs between scenarios with NVP risk multipliers of 5% vs. 40% is calculated as (Averted SADs with 40% NVP risk - Averted SADs with 5% NVP risk) / Averted SADs with 5% NVP risk.
^e The initial values for each input parameters in NVP Scenario are as following. NVP switching rate ( ${\sigma }_{a}$ ) = Males (Females): 4% (2.5%) for ages 24 and below, 2.5% (2.0%) for ages 25–34, 2.5% (1.6%) for age 35–44, 1.3% (1.4%) for ages 45–54, 1.2% (1.4%) for ages 55–64, and 0.6% (1.0%) for ages 65 and above; ${\alpha }_{S}$ = 75%; ${\alpha }_{V}$ = 50%; ${\chi }_{S}$ =100%; ${\chi }_{V}$ =100%; $\delta =0\%.$
^f NVP switching rate is the annual rate at which current smokers switch to NVPs.
^g Annual decay rate is the exponential rate of decline in switching rates over time.
^h Smoking initiation multiplier is relative to smoking initiation in the No-NVP Scenario.
ⁱ NVP initiation multiplier is relative to smoking initiation rates in the No-NVP Scenario.
^j Smoking cessation multiplier is relative to smoking cessation in the No-NVP Scenario.
^k NVP cessation multiplier is relative to smoking cessation rates in the No-NVP Scenario.

Table 6

Sensitivity analysis: Life-years lost and averted life years lost in the No-NVP Scenario and NVP Scenario across parameter changes, both genders, ages 18–99, 2013–2100
Scenario	NVP relative risk (Risk_NVP)^a = 5%		NVP relative risk (Risk_NVP) = 40%		Relative Change (5% vs 40%)^d
No-NVP Scenario	Total LYLs		Total LYLs		Relative Change (5% vs 40%)^d
	298,388,000	-	298,388,000	-
NVP Scenario with parameter changes from baseline	Averted LYLs^b	Relative Change (vs. baseline estimate)^c	Averted LYLs^b	Relative Change (vs. baseline estimate)^c
Baseline Estimate^e	97,811,967	-	51,904,506	-	-46.9%
50% of switch rate (σ_a),^f no decay (δ = 0%) ^g	70,025,600	-28.4%	34,869,552	-32.8%	-50.2%
200% of switch rate (σ_a), no decay (δ=0%)	129,004,951	31.9%	71,011,990	36.8%	-45.0%
100% of switch rate (σ_a), decay (δ= −0.20%)	36,052,459	-63.1%	13,960,932	-73.1%	-61.3%
25% of smoking initiation multiplier (α_S)^h	118,118,260	20.8%	81,026,822	56.1%	-31.4%
125% of smoking initiation multiplier (α_S)	80,006,203	-18.2%	26,273,401	-49.4%	-67.2%
25% of NVP initiation (α_V)ⁱ	97,981,994	0.2%	60,046,621	15.7%	-38.7%
75% of NVP initiation (α_V)	97,672,846	-0.1%	44,283,708	-14.7%	-54.7%
50% of smoking cessation (χ_S)^j	64,946,262	-33.6%	13,426,132	-74.1%	-79.3%
150% of smoking cessation (χ_S)	117,770,436	20.4%	75,901,585	46.2%	-35.6%
50% of NVP cessation (χ_V)^k	94,648,016	-3.2%	28,084,183	-45.9%	-70.3%
150% of NVP cessation (χ_V)	99,516,947	1.7%	64,898,048	25.0%	-34.8%
Notes: NVP = nicotine vaping product, LYLs = life-years lost
^a The NVP relative risk multiplier is the mortality risk of NVPs as a percentage of the excess
mortality risk of smoking.
^b The absolute reduction in life-years lost in the NVP Scenario compared to the No-NVP Scenario over 2013–2100.
^c The relative percent change in averted LYLs for each NVP Scenario compared to the initial NVP Scenario (best estimate). A negative (positive) value means that changing the parameter will decrease (increase) the averted LYLs in the specific scenario relative to averted LYLs in the initial NVP Scenario.
^d The relative percent change in averted LYLs between scenarios with NVP risk multipliers of 5% vs. 40% is calculated as (Averted LYLs with 40% NVP risk - Averted LYLs with 5% NVP risk) / Averted LYLs with 5% NVP risk.
^e The initial values for each input parameters in NVP Scenario are as following. NVP switching rate ( ${\sigma }_{a}$ ) = Males (Females): 4% (2.5%) for ages 24 and below, 2.5% (2.0%) for ages 25–34, 2.5% (1.6%) for age 35–44, 1.3% (1.4%) for ages 45–54, 1.2% (1.4%) for ages 55–64, and 0.6% (1.0%) for ages 65 and above; ${\alpha }_{S}$ = 75%; ${\alpha }_{V}$ = 50%; ${\chi }_{S}$ =100%; ${\chi }_{V}$ =100%; $\delta =0\%.$
^f NVP switching rate is the annual rate at which current smokers switch to NVPs.
^g Annual decay rate is the exponential rate of decline in switching rates over time.
^h Smoking initiation multiplier is relative to smoking initiation in the No-NVP Scenario.
ⁱ NVP initiation multiplier is relative to smoking initiation rates in the No-NVP Scenario.
^j Smoking cessation multiplier is relative to smoking cessation in the No-NVP Scenario.
^k NVP cessation multiplier is relative to smoking cessation rates in the No-NVP Scenario.

Tables 5 and 6 here

If the baseline NVP switching rate parameter (σ_a) is reduced by 50% with 0% decay (δ = 0%), NVP replacement of smoking is less in the NVP Scenario, and averted SVADs decline in relative terms by 28.8% from 5.9 million to 4.2 million; averted LYLs decline by 28.4% from 97.8 million to 70.0 million. A doubling of the NVP switching rates from their initial levels from PATH yields a 34.4% relative increase in averted SVADs to 7.9 million and a 31.9% relative increase in averted LYLs to 129.0 million. Using the initial NVP switching rates but with a 20% decay rate (δ= -20%) reduces averted SVADs by 60.4% to 2.3 million and reduces averted LYLs by 63.1% to 36.1 million.

Increasing the smoking initiation multiplier (α_S) from its baseline level of 75% primarily increases youth and young adult use at first, leading to more SVADs and LYLs later in life. An increase from 75–125% yields 4.6 million averted SVADs, a relative reduction of 22.1% compared to the initial NVP Scenario, and 80.0 million averted LYLs, a relative reduction of 18.2%. With the smoking initiation multiplier at 25%, averted SVADs increase to 7.4 million, a relative increase of 24.8%, and averted LYLs increase to 118.1 million, a relative increase 20.8%. Increasing or decreasing the NVP initiation multiplier (α_V) leads to a much smaller relative change in averted SVADs and LYLs due to the lower risks of NVPs relative to smoking (5%). Increasing the NVP initiation multiplier from 50–75% yields a relative reduction of 0.16% in averted SVADs and of 0.14% in averted LYLs. A reduction in the NVP initiation multiplier to 25% resulted in the increase in averted SVADs by 0.19% and in averted LYLs by 0.17%.

Compared to changes in the initiation parameters, changes in cessation multipliers lead to a more immediate impact on SVADs, since cessation generally occurs later in life when mortality risk is highest. Increasing the smoking cessation multiplier (χ_S) from 100–150% of cessation in the No-NVP Scenario yields a 20.2% relative increase to 7.1 million averted SVADs and a 20.4% relative increase to 117.8 million averted LYLs. Reducing the smoking cessation multiplier from 100–50% yields a 38.6% reduction to 3.6 million averted SVADs and a 33.6% reduction to 64.9 million averted LYLs. Comparable changes in the NVP cessation parameters yield less change. Increasing the NVP cessation multiplier (χ_V) from 100–150%, averted SVADs increased by 1.8% and LYLs increased by 1.7%. Reducing the NVP cessation multiplier to 50%, averted SVADs and LYLs decline by 4.3% and 3.2%, respectively.

Thus, increasing NVP risks leads to greater sensitivity of LYLs to changes in switching, NVP initiation, and NVP cessation parameters, but still yields public health gains. Since the NVP risk is substantially lower than the smoking risk, the relative changes in averted SVADs and LYLs using the NVP initiation or cessation multipliers are less than the relative changes using the smoking initiation or cessation multipliers for comparable changes.

Sensitivity to NVP Relative Risks

Tables 5 and 6 provide estimates of averted SVADs and LYLs with an NVP relative risk multiplier (Risk_NVP) of 40% (columns 4 and 5) instead of the baseline 5%, under the different NVP transition parameter ranges. With all other parameters at their baseline levels in the NVP Scenario, averted SVADs decrease in relative terms by 50.0% from 5.9 million with 5% NVP risks to 2.9 million with 40% NVP risks, and averted LYLs decrease by 46.9% from 97.8 million to 51.9 million.

Compared to an NVP relative risk of 5%, an NVP relative risk of 40% yields a 50.2% relative reduction in averted LYLs when reducing the NVP switching rates (σ_a) by 50%, while increasing the NVP switching rates by 100% yields a 45.0% relative reduction compared to 5% NVP risk. Reducing the smoking initiation multiplier (α_S) from 75–25% yields a 31.4% relative reduction in LYLs, while an increase to 125% yields a 67.2% relative reduction. Reducing the NVP initiation multiplier (α_V) from 50–25% yields 38.7% fewer averted LYLs, while an increase to 75% yields 54.7% fewer averted LYLs. Reducing the smoking cessation multiplier (χ_S) from 100–50% yields 79.3% fewer averted LYLs, while an increase to 150% yields 35.6% fewer averted LYLs. Reducing the NVP cessation rate multiplier (χ) from 100–50% yields 70.3% fewer averted LYLs, while an increase to 150% yields 34.8% fewer averted LYLs.

The SAVM is among the first validated simulation models incorporating vaping. SAVM estimates for smoking prevalence trends were generally close to those from 2013–2018 NHIS observational data, except at younger ages where reductions in smoking prevalence in recent years were greater than those predicted by the model. The validation of the 2018 NVP use estimate was less clear due to our assumption that NVP use begins at a zero level in 2013, but SAVM 2018 estimates were generally consistent with those of the 2018 NHIS.

Based on our initial estimates of NVP health risk, switching rates, smoking and NVP initiation rates, and smoking and NVP cessation rates, the SAVM projects that NVP use in the population will translate to 5.9 million premature smoking- and vaping-attributable deaths avoided and 97.8 million life years gained from 2013–2100. Our sensitivity analyses showed how different parameters influence the estimated impact of NVP use on public health.

There is a considerable uncertainty and controversy surrounding the relative risks of NVP use compared to smoking (Eissenberg et al. 2020). The initial NVP relative risk multiplier was set to 5% (Nutt et al. 2016,McNeill et al. March 2020), but when increased to 40%, the associated public health gains were substantially reduced, although still positive.

Model estimates are particularly sensitive to the rate of switching from smoking to NVP use and smoking initiation in the NVP Scenario. These rates, particularly the NVP switching rates of young adults, may explain the failure of the SAVM to predict the recent rapid decline in smoking by those ages 18–24. Recent studies (Al Rifai et al. 2020, Creamer et al. 2019, Levy et al. 2019b, Mirbolouk et al. 2018, Vallone et al. 2020) indicate that NVP use among US young adult smokers may be higher and smoking rates lower than indicated by our switching rates. We assume that NVPs lead to a net decrease in smoking initiation rates (i.e., α_S < 1). However, the downward trend in youth and young adult smoking has been especially great since 2017, at the same time vaping devices such as JUUL gained popularity (Chu et al. 2018, Jackler and Ramamurthi 2019, McKeganey and Russell 2019, Vallone et al. 2018). Youth and young adult smoking prevalence dropped by as much as 40% relative to earlier trends at about the time when NVP use increased in these same age groups (Levy et al. 2019a) and the 2019 Monitoring the Future survey (University of Michigan 2019) indicates past 30 day smoking rates as low as 5.8% and daily smoking rates as low as 2.4%. This recent data suggests that our smoking initiation multiplier parameter. The role of NVPs in the initiation of smoking is particularly controversial (Levy et al. 2017), but, like earlier work(Cherng et al. 2016, National Academy of Sciences 2018, Warner and Mendez 2019), public health impacts are relatively insensitive to the NVP initiation parameter. Nevertheless, while there is currently considerable uncertainty about NVP initiation, especially regarding more regular use (Hammond et al. 2019, Miech et al. 2019a, Miech et al. 2019b, Miech et al. 2019c).

Public health estimates are sensitive to the smoking cessation parameter in the NVP Scenario. As a conservative estimate, we assume that the smoking cessation multiplier was 100%, but cessation may be higher than in the No-NVP Scenario to the extent that NVPs are an effective smoking cessation alternative (Hajek, McRobbie and Bullen 2016, Przulj, Hajek and Phillips-Waller 2019, Zhu et al. 2017, Zhuang et al. 2016). Public health estimates are less sensitive to the NVP cessation multiplier when the NVP relative risks are 5%, but more sensitive when the risk increases to 40%. Our NVP cessation multiplier baseline value of 100% assumes NVP cessation rates are identical to that of smoking, but some recent evidence indicates that NVPs may be less addictive than smoking, implying a value greater than 100% (Bold et al. 2018, Liu et al. 2017, Martinez et al. 2019, Morean, Krishnan-Sarin and O'Malley 2018, Strong et al. 2017). These rates, along with the NVP and smoking initiation parameters, may vary over time (e.g., due to technological improvements in NVP or because those most susceptible to quitting have already quit) with age- and gender-related variations.

Government regulations could dramatically influence the uptake, use, and risks of NVPs. (e.g., by the FDA in the US). If the health risks of using NVPs are found to be higher than anticipated, those risks may be reduced with prudent FDA regulations that ban or limit the presence of known toxicants in e-liquids and by implementing product standards that encourage future NVP industry innovations that reduce NVP risks. Product regulations can also reduce the addictiveness of NVPs by reducing allowable nicotine content, thereby reducing NVP initiation and increasing NVP cessation. Policies that make NVPs more desirable relative to smoking could improve public health through increased switching from cigarettes to NVPs and increased cessation from smoking. Finally, the public health benefits of product regulation and industry innovation can be amplified with strong policies that continue to be directed at smoking—which would improve public health through reduced smoking initiation and increased smoking cessation. In particular, higher cigarette taxes relative to NVP taxes, e.g., in proportion to their risks, is likely to increase switching from cigarette to NVP use (Chaloupka, Sweanor and Warner 2015, Huang et al. 2018, Pesko et al. 2018, Stoklosa, Drope and Chaloupka 2016, Zheng et al. 2016).

Our analysis is subject to limitations. Future projections in the No-NVP Scenario are based on data through 2013 and we assumed that the projected trends would have simply continued. While our No-NVP Scenario is validated based on trends informed by data through 2013, future trends are subject to uncertainty and would depend on tobacco control policies and other environmental changes that would have occurred if NVPs had not come onto the US market. Our validation of smoking prevalence indicates that the model under-predicts reductions for those ages 18–24 and over-predicts reductions for those ages 65 and above. Further analysis is warranted. Our validation of NVP use is sensitive to the measure of NVP use. We applied a measure of at least 10 of the last 30 days as a definition of regular use, but measures of regular use are not well defined. Measures of everyday/someday use and at least 1 day in the past 30 days yield higher estimates, especially at younger ages.

The SAVM applies simplifying assumptions so that the model is user-friendly and can be adapted to individual needs. In the SAVM, a never smoker who temporarily uses NVPs and does not transition to smoking is considered a non-user. Our analysis of public health impacts also depends on the smoking prevalence projected in the absence of NVPs. However, the model validated well through the year 2018. Because transitions to and from dual use are often particularly unstable (Azagba, Shan and Latham 2019, Borland et al. 2019, Coleman et al. 2018, Robertson et al. 2019), we have chosen not to distinguish dual use from cigarette smoking in the NVP Scenario. In addition, the model does not consider relapse, so transitions to former smoker or former NVP users are treated as permanent, consistent with previous CISNET models (Holford et al. 2014a, Holford et al. 2014b, Holford, Levy and Meza 2016). Our analysis also does not explicitly consider smokers who temporarily use NVPs (as a dual user or exclusive user) and quickly transition to no cigarette and no NVP as a former smoker.

Except for the switching parameter, we have assumed that the other NVP risk and transitions parameters are age and time invariant, because current information was considered insufficient to indicate otherwise. As indicated above, these parameters are likely to depend on regulatory policies toward NVPs. As more information becomes available, it will be important to integrate such variation in key parameters.

In general, the relative risk, switching, initiation, and cessation parameters, while essential to the model, are subject to considerable uncertainty. Compared to smoking, NVPs are new products with less empirical evidence on their long-term health effects and usage/transition patterns (Fairchild, Bayer and Lee 2019, Green, Bayer and Fairchild 2016). In addition, patterns of future NVP use are difficult to predict. NVP use is still relatively new and use patterns can be expected to change as a result of the “disruptive” nature of the product (Abrams 2014, Levy et al. 2019c). Future use will depend on product innovations and regulations surrounding NVP content, marketing and use.

Although other categories of tobacco products are available on the market, including cigars, smokeless tobacco, and heated tobacco products (HTPs) (Czoli et al. 2020), we only consider two categories of products, cigarettes and NVPs, to keep the model tractable. However, the model is flexible enough to simulate patterns of switching and substitution between cigarettes and any other category of nicotine delivery products. If users wish to combine HTPs and NVPs into the same category as reduced harm alternative products, the model in its current structure would not be able to account for complex patterns of substitution between cigarettes, NVPs and HTPs.

The SAVM model has been developed in such a way that it can be easily modified. The model is developed in Excel both for transparency and to ensure its accessibility to non-technical audiences. More experienced users can modify the model’s underlying assumptions and structure as described in the SAVM User Guide, thereby providing the user flexibility to gauge the public health impact of NVP use and the potential impact of policies directed at NVP and cigarette use. The model and User Guide are readily available.

In conclusion, the SAVM was developed as a user-friendly tool to examine the potential public health impact of NVP use on smoking behaviors and to show how public health outcomes vary across different assumptions related to NVP health risks and the initiation and cessation of NVP use and smoking. The analysis indicates that under current assumptions and best estimates for model parameters, NVP use in the US would be associated with major public health gains going into the future. Using readily available data, SAVM can be applied by policymakers, researchers and other stakeholders in other countries to help understand the role of NVPs vis’ a vis’ smoking and the impact that those products have on public health.

NVP = nicotine vaping product, SAVM = Smoking and Vaping Model. SADs = smoking-attributable deaths, LYLs = life years lost.

Ethics approval and consent to participate: All original data are publicly available.

Consent for publication: All authors have approved the manuscript for submission.

Availability of data and materials: The model, data used therein and a 100 page User Manual will be made available to all requestors, and we will make the model available on our FDA Tobacco Center of Regulatory Science (TCORS) website.

Competing interests: We have no conflicts of interests.

Funding: Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health (NIH) and FDA Center for Tobacco Products (CTP) under award number U54CA229974. Support was also received for Dr. Levy from the National Cancer Institute under award number P01 CA200512. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the Food and Drug Administration.

Authors' contributions: DTL wrote the original draft. ZY and YL did the initial modeling and helped write the original methods section. All authors contributed to the development of the model and have contributed to the writing of the paper.

Acknowledgements: We would like to thank Ted Holford for comments on an earlier draft of the paper.

Abrams DB. "Promise and Peril of E-Cigarettes: Can Disruptive Technology Make Cigarettes Obsolete?". JAMA. 2014;311(2):135–6. doi:10.1001/jama.2013.285347.
Al Rifai M, Merchant AT, Nambi V, Jia X, Gulati M, Valero-Elizondo J, Nasir K, Ballantyne CM, Virani SS. "Temporal Trends in E-Cigarette Use among U.S. Adults: Behavioral Risk Factor Surveillance System, 2016 to 2018.". Am J Med. 2020. doi:10.1016/j.amjmed.2019.12.020.
Azagba S, Shan L, Latham K. "Adolescent Dual Use Classification and Its Association with Nicotine Dependence and Quit Intentions.". J Adolesc Health. 2019;65(2):195–201. doi:10.1016/j.jadohealth.2019.04.009.
Bachand AM, Sulsky SI, Curtin GM. "Assessing the Likelihood and Magnitude of a Population Health Benefit Following the Market Introduction of a Modified-Risk Tobacco Product: Enhancements to the Dynamic Population Modeler, Dpm(+ 1).". Risk Anal. 2018;38(1):151–62. doi:10.1111/risa.12819.
Bold KW, Sussman S, O'Malley SS, Grana R, Foulds J, Fishbein H, Krishnan-Sarin S. Measuring E-Cigarette Dependence: Initial Guidance. Addict Behav. 2018;79:213–18. doi:10.1016/j.addbeh.2017.11.015.
10.1111/add.14570
Borland R, Murray K, Gravely S, Fong GT, Thompson ME, McNeill A, O'Connor RJ, Goniewicz ML, Yong HH, Levy DT, Heckman BW, Cummings KM. 2019. "A New Classification System for Describing Concurrent Use of Nicotine Vaping Products Alongside Cigarettes (So-Called 'Dual Use'): Findings from the Itc-4 Country Smoking and Vaping Wave 1 Survey." Addiction. doi: 10.1111/add.14570.
Bullen C, Howe C, Laugesen M, McRobbie H, Parag V, Williman J, Walker N. Electronic Cigarettes for Smoking Cessation: A Randomised Controlled Trial. Lancet. 2013;382(9905):1629–37. doi:10.1016/S0140-6736(13)61842-5.
Caraballo RS, Shafer PR, Patel D, Davis KC, McAfee TA. Quit Methods Used by Us Adult Cigarette Smokers, 2014–2016. Prev Chronic Dis. 2017;14:E32. doi:10.5888/pcd14.160600.
Carter BD, Abnet CC, Feskanich D, Freedman ND, Hartge P, Lewis CE, Ockene JK, Prentice RL, Speizer FE, Thun MJ, Jacobs EJ. "Smoking and Mortality–Beyond Established Causes.". N Engl J Med. 2015;372(7):631–40. doi:10.1056/NEJMsa1407211.
Center for Disease Control and Prevention. 2019, "National Center for Health Statistics, Population, Deaths and Lifeyear Lostby Age and Gender". Retrieved November 15, 2019.
Centers for Disease Control and Prevention. 2019a, "Us Population and Mortality Data". Retrieved November 1, 2019.
Centers for Disease Control and Prevention,. 2019b. "National Health Interview Survey." Vol. 2019.
Chaloupka FJ, Sweanor D, Warner KE. Differential Taxes for Differential Risks–toward Reduced Harm from Nicotine-Yielding Products. N Engl J Med. 2015;373(7):594–7. doi:10.1056/NEJMp1505710.
Cherng ST, Tam J, Christine PJ, Meza R. Modeling the Effects of E-Cigarettes on Smoking Behavior: Implications for Future Adult Smoking Prevalence. Epidemiology. 2016;27(6):819–26. doi:10.1097/EDE.0000000000000497.
Chu KH, Colditz JB, Primack BA, Shensa A, Allem JP, Miller E, Unger JB, Cruz TB. "Juul: Spreading Online and Offline.". J Adolesc Health. 2018;63(5):582–86. doi:10.1016/j.jadohealth.2018.08.002.
CISNET. 2016, "Smoking Histories". Retrieved December 20, 2019 https://resources.cisnet.cancer.gov/projects/.
Cobb CO, Villanti AC, Graham AL, Pearson JL, Glasser AM, Rath JM, Stanton CA, Levy DT, Abrams DB, Niaura R. Markov Modeling to Estimate the Population Impact of Emerging Tobacco Products: A Proof-of-Concept Study. Tobacco Regulatory Science. 2015;1(2):129–41.
10.1136/tobaccocontrol-2017-054174
Coleman B, Rostron B, Johnson SE, Persoskie A, Pearson J, Stanton C, Choi K, Anic G, Goniewicz ML, Cummings KM, Kasza KA, Silveira ML, Delnevo C, Niaura R, Abrams DB, Kimmel HL, Borek N, Compton WM, Hyland A. 2018. "Transitions in Electronic Cigarette Use among Adults in the Population Assessment of Tobacco and Health (Path) Study, Waves 1 and 2 (2013–2015)." Tob Control. doi: 10.1136/tobaccocontrol-2017-054174.
Creamer MR, Wang TW, Babb S, Cullen KA, Day H, Willis G, Jamal A, Neff L. Tobacco Product Use and Cessation Indicators among Adults - United States, 2018. MMWR Morb Mortal Wkly Rep. 2019;68(45):1013–19. doi:10.15585/mmwr.mm6845a2.
10.1001/jama.2019.18387
Cullen KA, Gentzke AS, Sawdey MD, Chang JT, Anic GM, Wang TW, Creamer MR, Jamal A, Ambrose BK and B. A. King. 2019. "E-Cigarette Use among Youth in the United States, 2019." JAMA. doi: 10.1001/jama.2019.18387.
Czoli CD, White CM, Reid JL, Connor O RJ and Hammond D. Awareness and Interest in Iqos Heated Tobacco Products among Youth in Canada, England and the USA. Tob Control. 2020;29(1):89–95. doi:10.1136/tobaccocontrol-2018-054654.
Dai H, Leventhal AM. 2019. "Prevalence of E-Cigarette Use among Adults in the United States, 2014–2018." JAMA. doi: 10.1001/jama.2019.15331.
de Groot PM, Wu CC, Carter BW, Munden RF. The Epidemiology of Lung Cancer. Transl Lung Cancer Res. 2018;7(3):220–33. doi:10.21037/tlcr.2018.05.06.
DeVito EE, Krishnan-Sarin S. E-Cigarettes: Impact of E-Liquid Components and Device Characteristics on Nicotine Exposure. Curr Neuropharmacol. 2018;16(4):438–59. doi:10.2174/1570159X15666171016164430.
Eissenberg T, Bhatnagar A, Chapman S, Jordt SE, Shihadeh A, Soule EK. "Invalidity of an Oft-Cited Estimate of the Relative Harms of Electronic Cigarettes.". Am J Public Health. 2020;110(2):161–62. doi:10.2105/AJPH.2019.305424.
Etter JF. "Throat Hit in Users of the Electronic Cigarette: An Exploratory Study.". Psychol Addict Behav. 2016;30(1):93–100. doi:10.1037/adb0000137.
Fairchild AL, Bayer R, Lee JS. "The E-Cigarette Debate: What Counts as Evidence?". Am J Public Health. 2019;109(7):1000–06. doi:10.2105/AJPH.2019.305107.
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. doi:10.1038/srep04133.
Federal Register,. 2014, " Deeming Tobacco Products to Be Subject to the Federal Food, Drug, and Cosmetic Act, as Amended by the Family Smoking Prevention and Tobacco Control Act; Regulations on the Sale and Distribution of Tobacco Products and Required Warning Statements for Tobacco Products". https://www.federalregister.gov/articles/2014/04/25/2014-09491/deeming-tobacco-products-to-be-subject-to-the-federal-food-drug-and-cosmetic-act-as-amended-by-the#h-39.
Filippidis FT, Laverty AA, Gerovasili V, Vardavas CI. Two-Year Trends and Predictors of E-Cigarette Use in 27 European Union Member States. Tob Control. 2017;26(1):98–104. doi:10.1136/tobaccocontrol-2015-052771.
Giovenco DP, Delnevo CD. Prevalence of Population Smoking Cessation by Electronic Cigarette Use Status in a National Sample of Recent Smokers. Addict Behav. 2018;76:129–34. doi:10.1016/j.addbeh.2017.08.002.
Glasser AM, Collins L, Pearson JL, Abudayyeh H, Niaura RS, Abrams DB, Villanti AC. "Overview of Electronic Nicotine Delivery Systems: A Systematic Review.". Am J Prev Med. 2016. doi:10.1016/j.amepre.2016.10.036.
Goldenson NI, Kirkpatrick MG, Barrington-Trimis JL, Pang RD, McBeth JF, Pentz MA, Samet JM, Leventhal AM. "Effects of Sweet Flavorings and Nicotine on the Appeal and Sensory Properties of E-Cigarettes among Young Adult Vapers: Application of a Novel Methodology.". Drug Alcohol Depend. 2016;168:176–80. doi:10.1016/j.drugalcdep.2016.09.014.
10.1136/tobaccocontrol-2012-050859
Goniewicz ML, Knysak J, Gawron M, Kosmider L, Sobczak A, Kurek J, Prokopowicz A, Jablonska-Czapla M, Rosik-Dulewska C, Havel C, Jacob P, 3rd and N. Benowitz. 2014. "Levels of Selected Carcinogens and Toxicants in Vapour from Electronic Cigarettes." Tob Control 23(2):133-9. doi: 10.1136/tobaccocontrol-2012-050859.
Goniewicz ML, Gawron M, Smith DM, Peng M, Jacob P, 3rd and Benowitz NL. 2017. "Exposure to Nicotine and Selected Toxicants in Cigarette Smokers Who Switched to Electronic Cigarettes: A Longitudinal within-Subjects Observational Study." Nicotine Tob Res 19(2):160–67. doi:10.1093/ntr/ntw160.
10.1056/NEJMc1606395
Green SH, Bayer R, Fairchild AL. 2016. "Evidence, Policy, and Cigarettes." E-. N Engl J Med 375(5):e6. doi: 10.1056/NEJMc1606395.
Hajek P, McRobbie H, Bullen C. "E-Cigarettes and Smoking Cessation.". Lancet Respir Med. 2016;4(6):e23. doi:10.1016/S2213-2600(16)30024-8.
Hajek P, Przulj D, Phillips-Waller A, Anderson R, McRobbie H. Initial Ratings of Different Types of E-Cigarettes and Relationships between Product Appeal and Nicotine Delivery. Psychopharmacology. 2018;235(4):1083–92. doi:10.1007/s00213-017-4826-z.
Hajek P, Phillips-Waller A, Przulj D, Pesola F, Myers Smith K, Bisal N, Li J, Parrott S, Sasieni P, Dawkins L, Ross L, Goniewicz M, Wu Q. and H. J. McRobbie. 2019. "A Randomized Trial of E-Cigarettes Versus Nicotine-Replacement Therapy." N Engl J Med 380(7):629–37. doi: 10.1056/NEJMoa1808779.
Hammond D, Reid JL, Rynard VL, Fong GT, Cummings KM, McNeill A, Hitchman S, Thrasher JF, Goniewicz ML, Bansal-Travers M, O'Connor R, Levy D, Borland R, White CM. Prevalence of Vaping and Smoking among Adolescents in Canada, England, and the United States: Repeat National Cross Sectional Surveys. BMJ. 2019;365:l2219. doi:10.1136/bmj.l2219.
Hecht SS, Carmella SG, Kotandeniya D, Pillsbury ME, Chen M, Ransom BW, Vogel RI, Thompson E, Murphy SE, Hatsukami DK. Evaluation of Toxicant and Carcinogen Metabolites in the Urine of E-Cigarette Users Versus Cigarette Smokers. Nicotine Tob Res. 2014. doi:10.1093/ntr/ntu218.
Hill A, Camacho OM. A System Dynamics Modelling Approach to Assess the Impact of Launching a New Nicotine Product on Population Health Outcomes. Regul Toxicol Pharmacol. 2017;86:265–78. doi:10.1016/j.yrtph.2017.03.012.
Holford TR, Levy DT, McKay LA, Clarke L, Racine B, Meza R, Land S, Jeon J, Feuer EJ. Patterns of Birth Cohort-Specific Smoking Histories, 1965–2009. Am J Prev Med. 2014a;46(2):e31-7. doi:10.1016/j.amepre.2013.10.022.
10.1001/jama.2013.285112
Holford TR, Meza R, Warner KE, Meernik C, Jeon J, Moolgavkar SH, Levy DT. 2014b. "Tobacco Control and the Reduction in Smoking-Related Premature Deaths in the United States, 1964–2012." JAMA 311(2):164 – 71. doi: 10.1001/jama.2013.285112.
Holford TR, Levy DT, Meza R. Comparison of Smoking History Patterns among African American and White Cohorts in the United States Born 1890 to 1990. Nicotine Tob Res. 2016;18(Suppl 1):16–29. doi:10.1093/ntr/ntv274.
Huang J, Gwarnicki C, Xu X, Caraballo RS, Wada R, Chaloupka FJ. "A Comprehensive Examination of Own- and Cross-Price Elasticities of Tobacco and Nicotine Replacement Products in the U.S.". Prev Med. 2018;117:107–14. doi:10.1016/j.ypmed.2018.04.024.
Jackler RK, Ramamurthi D. Nicotine Arms Race: Juul and the High-Nicotine Product Market. Tob Control. 2019. doi:10.1136/tobaccocontrol-2018-054796.
Jeon J, Holford TR, Levy DT, Feuer EJ, Cao P, Tam J, Clarke L, Clarke J, Kong CY, Meza R. "Smoking and Lung Cancer Mortality in the United States from 2015 to 2065: A Comparative Modeling Approach.". Ann Intern Med. 2018;169(10):684–93. doi:10.7326/M18-1250.
Jha P, Ramasundarahettige C, Landsman V, Rostron B, Thun M, Anderson RN, McAfee T, Peto R. "21st-Century Hazards of Smoking and Benefits of Cessation in the United States.". N Engl J Med. 2013;368(4):341–50. doi:10.1056/NEJMsa1211128.
Kalkhoran S, Glantz SA. "Modeling the Health Effects of Expanding E-Cigarette Sales in the United States and United Kingdom: A Monte Carlo Analysis.". JAMA Intern Med. 2015;175(10):1671–80. doi:10.1001/jamainternmed.2015.4209.
Levy DT, Meza R, Zhang Y, Holford TR. "Gauging the Effect of U.S. Tobacco Control Policies from 1965 through 2014 Using Simsmoke.". Am J Prev Med. 2016;50(4):535–42. doi:10.1016/j.amepre.2015.10.001.
Levy DT, Cummings KM, Villanti AC, Niaura R, Abrams DB, Fong GT, Borland R. A Framework for Evaluating the Public Health Impact of E-Cigarettes and Other Vaporized Nicotine Products. Addiction. 2017;112(1):8–17. doi:10.1111/add.13394.
Levy DT, Yuan Z, Li Y. 2017. "The Prevalence and Characteristics of E-Cigarette Users in the U.S." Int J Environ Res Public Health 14(10). doi:10.3390/ijerph14101200.
Levy DT, Borland R, Lindblom EN, Goniewicz ML, Meza R, Holford TR, Yuan Z, Luo Y, O'Connor RJ, Niaura R, Abrams DB. "Potential Deaths Averted in USA by Replacing Cigarettes with E-Cigarettes.". Tob Control. 2018a;27(1):18–25. doi:10.1136/tobaccocontrol-2017-053759.
Levy DT, Tam J, Kuo C, Fong GT, Chaloupka F. The Impact of Implementing Tobacco Control Policies: The 2017 Tobacco Control Policy Scorecard. J Public Health Manag Pract. 2018b;24(5):448–57. doi:10.1097/PHH.0000000000000780.
Levy DT, Warner KE, Cummings KM, Hammond D, Kuo C, Fong GT, Thrasher JF, Goniewicz ML, Borland R. 2018c. "Examining the Relationship of Vaping to Smoking Initiation among US Youth and Young Adults: A Reality Check." Tob Control. doi: 10.1136/tobaccocontrol-2018-054446.
Levy DT, Yuan Z, Luo Y, Abrams DB. "The Relationship of E-Cigarette Use to Cigarette Quit Attempts and Cessation: Insights from a Large, Nationally Representative U.S. Survey.". Nicotine Tob Res. 2018d;20(8):931–39. doi:10.1093/ntr/ntx166.
Levy DT, Warner KE, Cummings KM, Hammond D, Kuo C, Fong GT, Thrasher JF, Goniewicz ML, Borland R. Examining the Relationship of Vaping to Smoking Initiation among Us Youth and Young Adults: A Reality Check. Tob Control. 2019a;28(6):629–35. doi:10.1136/tobaccocontrol-2018-054446.
Levy DT, Yuan Z, Li Y, Mays D, Sanchez-Romero LM. 2019b. "An Examination of the Variation in Estimates of E-Cigarette Prevalence among U.S. Adults." Int J Environ Res Public Health 16(17). doi:10.3390/ijerph16173164.
Levy DT, Lindblom E, Sweanor D, Chalpouka F, O'Connor R, Shang C, Palley T, Fong GT, Cummings KM, Goniewicz ML, Borland B. An Economic Analysis of the Pre-Deeming Us Market for Nicotine Vaping Products. Tobacco Regulatory Science. 2019c;5(2):169–81. doi:https://doi.org/10.18001/TRS.5.2.8.
Liu G, Wasserman E, Kong L, Foulds J. A Comparison of Nicotine Dependence among Exclusive E-Cigarette and Cigarette Users in the Path Study. Prev Med. 2017;104:86–91. doi:10.1016/j.ypmed.2017.04.001.
Martinez U, Martinez-Loredo V, Simmons VN, Meltzer LR, Drobes DJ, Brandon KO, Palmer AM, Eissenberg T, Bullen CR, Harrell PT, Brandon TH. "How Does Smoking and Nicotine Dependence Change after Onset of Vaping? A Retrospective Analysis of Dual Users.". Nicotine Tob Res. 2019. doi:10.1093/ntr/ntz043.
McKeganey N, Russell C. "Prevalence of Awareness and Use of Juul E-Cigarettes in a National Probability Sample of Adolescents in the United States.". Am J Health Behav. 2019;43(3):591–605. doi:10.5993/AJHB.43.3.13.
McNeill A, Brose L, Calder R, Bauld L, Robson D. "Vaping in England: An Evidence Update Including Mental Health and Pregnancy, March 2020: A Report Commissioned. by Public Health England" Public Health England Retrieved April. March 2020;3:2020. https://www.gov.uk/government/publications/vaping-in-england-evidence-update-march-2020.
Miech R, Johnston LD, O'Malley PM, Terry-McElrath YM. "The National Prevalence of Adolescent Nicotine Use in 2017: Estimates Taking into Account Student Reports of Substances Vaped.". Addict Behav Rep. 2019a;9:100159. doi:10.1016/j.abrep.2019.100159.
10.1056/NEJMc1910739
Miech R, Johnston L, O'Malley PM, Bachman JG, Patrick ME. 2019b. "Trends in Adolescent Vaping, 2017–2019." N Engl J Med 381(15):1490-91. doi: 10.1056/NEJMc1910739.
Miech R, Johnston L, O'Malley PM, Bachman JG, Patrick ME. "Adolescent Vaping and Nicotine Use in 2017–2018 - U.S. National Estimates.". N Engl J Med. 2019c;380(2):192–93. doi:10.1056/NEJMc1814130.
Mirbolouk M, Charkhchi P, Kianoush S, Uddin SMI, Orimoloye OA, Jaber R, Bhatnagar A, Benjamin EJ, Hall ME, DeFilippis AP, Maziak W, Nasir K, Blaha MJ. "Prevalence and Distribution of E-Cigarette Use among U.S. Adults: Behavioral Risk Factor Surveillance System, 2016.". Ann Intern Med. 2018;169(7):429–38. doi:10.7326/M17-3440.
Morean M, Krishnan-Sarin S, O'Malley SS. "Comparing Cigarette and E-Cigarette Dependence and Predicting Frequency of Smoking and E-Cigarette Use in Dual-Users of Cigarettes and E-Cigarettes.". Addict Behav. 2018;87:92–6. doi:10.1016/j.addbeh.2018.06.027.
National Academy of Sciences, Engineering and Medicine. 2018. Public Health Consequences of E-CigarettesCongress.
National Center for Health Statistics,. 2019, "Percentage of Adults Aged 18–24 Years Who Currently Smoke Cigarettes or Who Currently Use Electronic Cigarettes, by Year — National Health Interview Survey, United States, 2014–2018", MMWR. Retrieved December 8, 2018 https://www.cdc.gov/mmwr/volumes/68/wr/mm6839a6.htm.
Nelluri B, Murphy K, Mookadam F, Mookadam M. The Current Literature Regarding the Cardiovascular Effects of Electronic Cigarettes. Future Cardiol. 2016;12(2):167–79. doi:10.2217/fca.15.83.
"E-Cigarettes Are

10.1016/S0140-6736(15)00253-6

Nutt DJ, Phillips LD, Balfour D, Curran HV, Dockrell M, Foulds J, Fagerstrom K, Letlape K, Polosa R, Ramsey J, Sweanor D. 2016. "E-Cigarettes Are Less Harmful Than Smoking" Lancet 387(10024):1160–2. doi:10.1016/S0140-6736(15)00253-6.
Pesko MF, Huang J, Johnston LD, Chaloupka FJ. E-Cigarette Price Sensitivity among Middle- and High-School Students: Evidence from Monitoring the Future. Addiction. 2018;113(5):896–906. doi:10.1111/add.14119.
Poland B, Teischinger F. "Population Modeling of Modified Risk Tobacco Products Accounting for Smoking Reduction and Gradual Transitions of Relative Risk.". Nicotine Tob Res. 2017;19(11):1277–83. doi:10.1093/ntr/ntx070.
Przulj D, Hajek P, Phillips-Waller A. "E-Cigarettes Versus Nicotine-Replacement Therapy for Smoking Cessation. Reply.". N Engl J Med. 2019;380(20):1974–75. doi:10.1056/NEJMc1903758.
Ratajczak A, Feleszko W, Smith DM, Goniewicz M. "How Close Are We to Definitively Identifying the Respiratory Health Effects of E-Cigarettes?". Expert Rev Respir Med. 2018;12(7):549–56. doi:10.1080/17476348.2018.1483724.
Robertson L, Hoek J, Blank ML, Richards R, Ling P, Popova L. Dual Use of Electronic Nicotine Delivery Systems (Ends) and Smoked Tobacco: A Qualitative Analysis. Tob Control. 2019;28(1):13–9. doi:10.1136/tobaccocontrol-2017-054070.
Rosenberg MA, Feuer EJ, Yu B, Sun J, Henley SJ, Shanks TG, Anderson CM, McMahon PM, Thun MJ, Burns DM. Chapter 3: Cohort Life Tables by Smoking Status, Removing Lung Cancer as a Cause of Death. Risk Anal. 2012;32(Suppl 1):25–38. doi:10.1111/j.1539-6924.2011.01662.x.
Royal College of Physicians,. 2016. "Nicotine without Smoke: Tobacco Harm Reduction." Vol. London: RCP.
Shahab L, Goniewicz ML, Blount BC, Brown J, McNeill A, Alwis KU, Feng J, Wang L, West R. Nicotine, Carcinogen, and Toxin Exposure in Long-Term E-Cigarette and Nicotine Replacement Therapy Users: A Cross-Sectional Study. Ann Intern Med. 2017a. doi:10.7326/M16-1107.
Shahab L, Goniewicz ML, Blount BC, Brown J, West R. "E-Cigarettes and Toxin Exposure.". Ann Intern Med. 2017b;167(7):525–26. doi:10.7326/L17-0315.
Soneji SS, Sung HY, Primack BA, Pierce JP, Sargent JD. Quantifying Population-Level Health Benefits and Harms of E-Cigarette Use in the United States. PLoS One. 2018;13(3):e0193328. doi:10.1371/journal.pone.0193328.
Stanton CA, Sharma E, Edwards KC, Halenar MJ, Taylor KA, Kasza KA, Day H, Anic G, Gardner LD, Hammad HT, Bansal-Travers M, Limpert J, Borek N, Kimmel HL, Compton WM, Hyland A. Longitudinal Transitions of Exclusive and Polytobacco Electronic Nicotine Delivery Systems (Ends) Use among Youth, Young Adults and Adults in the USA: Findings from the Path Study Waves 1–3 (2013–2016). Tob Control. 2020;29(Suppl 3):s147–54. doi:10.1136/tobaccocontrol-2019-055574.
Stoklosa M, Drope J, Chaloupka FJ. Prices and E-Cigarette Demand: Evidence from the European Union. Nicotine Tob Res. 2016;18(10):1973–80. doi:10.1093/ntr/ntw109.
Strong DR, Pearson J, Ehlke S, Kirchner T, Abrams D, Taylor K, Compton WM, Conway KP, Lambert E, Green VR, Hull LC, Evans SE, Cummings KM, Goniewicz M, Hyland A, Niaura R. "Indicators of Dependence for Different Types of Tobacco Product Users: Descriptive Findings from Wave 1 (2013–2014) of the Population Assessment of Tobacco and Health (Path) Study.". Drug Alcohol Depend. 2017;178:257–66. doi:10.1016/j.drugalcdep.2017.05.010.
Tam J, Levy DT, Jeon J, Clarke J, Gilkeson S, Hall T, Feuer EJ, Holford TR, Meza R. "Projecting the Effects of Tobacco Control Policies in the USA through Microsimulation: A Study Protocol.". BMJ Open. 2018;8(3):e019169. doi:10.1136/bmjopen-2017-019169.
Taylor KA, Sharma E, Edwards KC, Halenar MJ, Kissin W, Kasza KA, Day H, Anic G, Gardner LD, Hammad HT, Hull LC, Bansal-Travers M, Limpert J, Borek N, Kimmel HL, Compton WM, Hyland A, Stanton C. Longitudinal Pathways of Exclusive and Polytobacco Cigarette Use among Youth, Young Adults and Adults in the USA: Findings from the Path Study Waves 1–3 (2013–2016). Tob Control. 2020;29(Suppl 3):s139–46. doi:10.1136/tobaccocontrol-2020-055630.
Tegin G, Mekala HM, Sarai SK, Lippmann S. "E-Cigarette Toxicity?". South Med J. 2018;111(1):35–8. doi:10.14423/SMJ.0000000000000749.
Thun MJ, Lopez AD, Hartge P. "Smoking-Related Mortality in the United States.". N Engl J Med. 2013;368(18):1753. doi:10.1056/NEJMc1302783.
U.S. Department of Health and Human Services,.. 2014. "The Health Consequences of Smoking—50 Years of Progress: A Report of the Surgeon General." Vol. Atlanta, GA.
Center for Tobacco Products
Food US, Administration D. Center for Tobacco Products. 2020, "Rules and Regulations". Retrieved May 22, 2020 https://www.fda.gov/tobacco-products/rules-regulations-and-guidance/rules-and-regulations.
Center for Tobacco Products, National Institutes of Health and National Institute on Drug Abuse
Food US, Administration D. Center for Tobacco Products, National Institutes of Health and National Institute on Drug Abuse. 2019. " Population Assessment of Tobacco and Health (Path) Study ". Washington, DC.
Unger M, Unger DW. E-Cigarettes/Electronic Nicotine Delivery Systems: A Word of Caution on Health and New Product Development. J Thorac Dis. 2018;10(Suppl 22):2588-S92. doi:10.21037/jtd.2018.07.99.
University of Michigan,. 2019, "Monitoring the Future Survey", Ann Arbor. Retrieved October 28, 2019 https://www.drugabuse.gov/trends-statistics/monitoring-future/monitoring-future-study-trends-in-prevalence-various-drugs.
Vallone DM, Bennett M, Xiao H, Pitzer L, Hair EC. Prevalence and Correlates of Juul Use among a National Sample of Youth and Young Adults. Tob Control. 2018. doi:10.1136/tobaccocontrol-2018-054693.
Vallone DM, Cuccia AF, Briggs J, Xiao H, Schillo BA, Hair EC.. .". "Electronic Cigarette and Juul Use among Adolescents and Young Adults. JAMA Pediatr. 2020. doi:10.1001/jamapediatrics.2019.5436.
10.1111/add.14602
Voos N, Kaiser L, Mahoney MC, Bradizza CM, Kozlowski LT, Benowitz NL, O'Connor RJ, Goniewicz ML. 2019. "Randomized within-Subject Trial to Evaluate Smokers' Initial Perceptions, Subjective Effects and Nicotine Delivery across Six Vaporized Nicotine Products." Addiction 114(7):1236-48. doi: 10.1111/add.14602.
Wagener TL, Floyd EL, Stepanov I, Driskill LM, Frank SG, Meier E, Leavens EL, Tackett AP, Molina N, Queimado L. Have Combustible Cigarettes Met Their Match? The Nicotine Delivery Profiles and Harmful Constituent Exposures of Second-Generation and Third-Generation Electronic Cigarette Users. Tob Control. 2016. doi:10.1136/tobaccocontrol-2016-053041.
Warner KE. "Tobacco Control Policies and Their Impacts. Past, Present, and Future.". Ann Am Thorac Soc. 2014;11(2):227–30. doi:10.1513/AnnalsATS.201307-244PS.
Warner KE, Sexton DW, Gillespie BW, Levy DT, Chaloupka FJ. Impact of Tobacco Control on Adult Per Capita Cigarette Consumption in the United States. Am J Public Health. 2014;104(1):83–9. doi:10.2105/AJPH.2013.301591.
Warner KE, Mendez D. E-Cigarettes: Comparing the Possible Risks of Increasing Smoking Initiation with the Potential Benefits of Increasing Smoking Cessation. Nicotine Tob Res. 2019;21(1):41–7. doi:10.1093/ntr/nty062.
Weitkunat R, Lee PN, Baker G, Sponsiello-Wang Z, Gonzalez-Zuloeta Ladd AM, Ludicke F. "A Novel Approach to Assess the Population Health Impact of Introducing a Modified Risk Tobacco Product.". Regul Toxicol Pharmacol. 2015;72(1):87–93. doi:10.1016/j.yrtph.2015.03.011.
Yingst JM, Foulds J, Veldheer S, Hrabovsky S, Trushin N, Eissenberg TT, Williams J, Richie JP, Nichols TT, Wilson SJ, Hobkirk AL. Nicotine Absorption During Electronic Cigarette Use among Regular Users. PLoS One. 2019;14(7):e0220300. doi:10.1371/journal.pone.0220300.
Zheng Y, Zhen C, Dench D, Nonnemaker JM. U.S. Demand for Tobacco Products in a System Framework. Health Econ. 2016. doi:10.1002/hec.3384.
Zhu SH, Zhuang YL, Wong S, Cummins SE, Tedeschi GJ. E-Cigarette Use and Associated Changes in Population Smoking Cessation: Evidence from Us Current Population Surveys. BMJ. 2017;358:j3262. doi:10.1136/bmj.j3262.
Zhuang YL, Cummins SE, Sun JY, Zhu SH. Long-Term E-Cigarette Use and Smoking Cessation: A Longitudinal Study with Us Population. Tob Control. 2016;25(Suppl 1):i90–5. doi:10.1136/tobaccocontrol-2016-053096.

Table 2. Smoking prevalence (%), validation of US SAVM against the NHIS, by age and gender,

in years 2013-2018

Age	Source	2013	2014	2015	2016	2017	2018	Relative difference 2013-2018
US Males
18+	SAVM	21.4	20.4	19.4	18.5	17.5	16.7	-22.2%
	NHIS	20.3	18.8	16.7	17.5	15.8	15.8	-22.2%
	95% CI	19.5-21.2	17.9-19.7	15.9-17.6	16.7-18.4	15.0-16.6	15.0-16.6
18-24	SAVM	19.9	18.2	16.6	15.3	14.2	13.3	-33.0%
	NHIS	21.6	18.6	15.2	14.6	12.1	8.5	-60.8%
	95% CI	18.6-24.6	15.9-21.4	12.5-18.0	11.9-17.2	9.7-14.5	6.4-10.5
25-44	SAVM	27.3	26.3	25.2	24.0	22.8	21.6	-21.0%
	NHIS	23.1	22.9	19.7	20.6	19.3	19.1	-17.2%
	95% CI	21.6-24.5	21.3-24.4	18.2-21.3	19.0-22.2	17.7-20.8	17.5-20.7
45-64	SAVM	20.4	19.5	18.6	17.8	17.1	16.3	-19.9%
	NHIS	21.6	19.3	17.8	19.4	17.3	18.3	-15.2%
	95% CI	20.2-23.1	17.7-20.9	16.5-19.2	18.0-20.8	15.9-18.6	16.9-19.7
65+	SAVM	12.2	11.9	11.5	11.0	10.6	10.2	-16.6%
	NHIS	10.6	9.7	9.8	10.2	9.0	9.9	-6.3%
	95% CI	9.3-12.0	8.4-11.0	8.5-11.1	8.9-11.5	7.9-10.2	8.7-11.1
US Females
18+	SAVM	15.9	15.2	14.5	13.8	13.2	12.6	-20.7%
	NHIS	15.4	14.9	13.5	13.6	12.1	12.0	-22.1%
	95% CI	14.7-16.1	14.1-15.7	12.8-14.2	12.9-14.2	11.5-12.8	11.3-12.6
18-24	SAVM	15.0	13.7	12.5	11.5	10.6	9.9	-33.7%
	NHIS	15.4	14.8	11.1	11.4	8.5	7.3	-52.6%
	95% CI	13.0-17.8	10.6-18.9	9.0-13.3	9.4-13.3	6.5-10.5	5.2-9.4
25-44	SAVM	20.6	19.9	19.2	18.5	17.7	16.9	-17.6%
	NHIS	17.0	17.4	15.6	14.7	12.9	14.2	-16.6%
	95% CI	15.8-18.2	16.1-18.6	14.4-16.9	13.6-15.9	11.8-14.1	12.9-15.5
45-64	SAVM	16.2	15.6	14.9	14.4	13.8	13.3	-18.1%
	NHIS	18.1	16.8	15.8	16.8	15.3	14.3	-20.9%
	95% CI	16.8-19.3	15.6-18.0	14.6-17.1	15.6-18.0	14.1-16.5	13.1-15.5
65+	SAVM	8.2	7.8	7.4	7.1	6.8	6.5	-20.3%
	NHIS	7.6	7.6	7.6	7.6	7.8	7.3	-4.0%
	95% CI	6.6-8.6	6.5-8.6	6.5-8.6	6.7-8.6	6.8-8.7	6.4-8.2

Notes: Current smokers in the NHIS (National Health Interview Survey) are those who have smoked at least 100 cigarettes lifetime, and currently smoke cigarettes some days or every day.

Table 3. Nicotine Vaping Product Prevalence (%), validation for US SAVM using NHIS, by

age and gender, in year 2018

US Males
Ages	18+	18-24	25-44	45-64	65+
SAVM	3.0	7.5	4.2	1.4	0.5
NHIS	3.1	6.8	4.7	1.5	0.6
NHIS 95% CI	2.7-3.5	4.8-8.8	3.8-5.6	1.0-1.9	0.3-0.9
US Females
Ages	18+	18-24	25-44	45-64	65+
SAVM	1.8	4.9	2.4	1.0	0.4
NHIS	1.5	3.1	1.9	1.3	0.5
NHIS 95% CI	1.3-1.8	1.8-4.5	1.4-2.3	1.0-1.7	0.3-0.7

Notes: CI= confidence interval. NVP users from NHIS (National Health Interview Survey) are those who use NVPs at least 10 of the past 30 days.

Download PDF

Journal Publication

published 17 Apr, 2021

Read the published version in Population Health Metrics →

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Public Health Implications of Vaping in the US: The Smoking and Vaping Simulation Model

Status:

Journal Publication

Version 1

Abstract

Background

Methods

Results

Discussion

Figures

Background

Methods

Model Structure

The No-NVP Scenario

The NVP Scenario

Public Health Outcomes

Data and Parameters

The No-NVP Scenario

The NVP Scenario

Model Validation and Analysis

Results

Validation of Smoking Prevalence

Validation of NVP Prevalence

Public Health Impact Under the Baseline NVP and No-NVP Scenario

Sensitivity to NVP Transition Parameters

Sensitivity to NVP Relative Risks

Discussion

Abbreviations

Declarations

References

Tables

Status:

Journal Publication

Version 1