Predictors of Quitting Dual Use of Electronic Cigarettes and Cigarettes During Pregnancy

Abstract

Background and Aims:

There is limited research surrounding dual maternal use of cigarettes and electronic cigarettes (e-cigarettes). We aimed to assess predictors of maternal quitting of cigarettes, e-cigarettes, and both during late pregnancy.

Materials and Methods:

We analyzed dual use (n = 4,006) and exclusive e-cigarette use (n = 1,685) among mothers using data from the 2016 to 2019 phase of the Pregnancy Risk Assessment Monitoring Systems (PRAMS), a nationally representative sample of the United States. Dual use and exclusive e-cigarette use were defined based on use reported during the 3 months before pregnancy and quitting was assessed during the last 3 months of pregnancy. Multinomial and binomial logistic regression models estimated the odds ratios and 95% confidence intervals for predictors of quitting status among mothers who reported dual use and exclusive e-cigarette use, respectively. Separate predictor analyses were conducted in the dual and exclusive e-cigarette use groups to see predictors of quitting e-cigarettes, cigarettes, or both.

Results:

The highest proportion of mothers who used cigarettes and e-cigarettes before pregnancy quit both during late pregnancy (46.2%), followed by those who quit e-cigarette use only (26.5%) and those who quit cigarette use only (6.6%). Among mothers who reported dual use, those who were African American or Asian, of Hispanic ethnicity, consumed alcohol before pregnancy, had higher education, were married, had diabetes, had higher annual household income, had nongovernmental health insurance, had more prenatal care visits, had a higher frequency of e-cigarette use before pregnancy, had a lower frequency of cigarette use before pregnancy, and smoked hookah around pregnancy had a higher likelihood of quitting both cigarette and e-cigarette use during late pregnancy.

Conclusions:

Quitting use of cigarettes and/or e-cigarettes was fairly common among mothers who reported dual use or e-cigarette use only. Sociodemographics, pregnancy characteristics, and use of other tobacco products predicted quitting use of both cigarettes and e-cigarettes during late pregnancy.

Introduction

Electronic cigarettes (“e-cigarettes”) and other electronic nicotine delivery systems are battery-operated devices that heat a liquid, usually composed of glycerin or propylene glycol as well as nicotine, flavors, and additives, to create an aerosol inhaled by the individual.¹ There has been a steady rise in the use of e-cigarettes, possibly due to perceived harm reduction² and advertisements.³ This may also apply to some pregnant women who perceive e-cigarettes as a safer alternative to combustible cigarettes (“cigarettes”) and thus switch from cigarettes to e-cigarettes.^4–6

Maternal cigarette use patterns before, during, and after pregnancy have been well studied, with majority of research agreeing that the prevalence of cigarette use was high before pregnancy, decreased sharply during pregnancy, and raised moderately after pregnancy.^7–11 Patterns of e-cigarette use among pregnant women are still being researched.

However, some work has been previously done in the Pregnancy Risk Assessment Monitoring System (PRAMS), which is an ongoing national survey of health and behaviors of women in the United States before, during, and after pregnancy.¹² A 2015 study using PRAMS data in two states (Oklahoma and Texas, N = 3,277) found that exclusive e-cigarette use followed a similar pattern to cigarette use surrounding pregnancy, with e-cigarette use at a high of 2.3% in the 3 months before pregnancy, at 0.5% during the last 3 months of pregnancy, and at 0.7% in the 2 to 6 months after pregnancy.⁸ Similarly, analyses of 38 states from PRAMS in 2016–2017 showed that 3.6% of women used e-cigarettes 3 months before pregnancy and 1.1% used e-cigarettes during the last 3 months of pregnancy.¹³

In 2020, it was reported that 8.1 million US adults who consumed tobacco used multiple products.¹⁴ As a result, there has been growing research to determine the health risks related to dual use of cigarettes and e-cigarettes, including among pregnant women. In a 2023 cross-sectional study of pregnant smokers, 80.3% of the study sample smoked conventional cigarettes, in addition to other alternative forms of tobacco such as waterpipes, straw cigarettes, e-cigarettes, and/or Bali cigarettes during pregnancy.¹⁵ In this study, investigators found that pregnant smokers reporting dual or poly-use of tobacco products during pregnancy were associated with a significantly higher prevalence of comorbidities, including cardiovascular, pulmonary, cancer, and other medical conditions, compared to pregnant smokers who used conventional cigarettes only.¹⁵ Furthermore, evidence of adverse birth outcomes among women who used e-cigarettes only during pregnancy has also been observed. For example, in a relatively small cohort of pregnant women, the risk of small-for-gestational-age births was higher among women who used e-cigarettes only, compared to those who reported no use.¹⁶

Although e-cigarettes were originally advertised as a smoking cessation aid, it has led to the dual use of cigarettes and e-cigarettes, creating new public health consequences. In one cohort of US adults who reported dual use at baseline, 53.5% remained dual use, 37.4% switched to exclusive cigarette use, 2.5% switched to exclusive e-cigarette use, and only 6.7% quit both at the 1-year follow-up.¹⁷ Another study conducted in 2015–2016 found that 55.6% of pregnant women who reported dual use stopped using e-cigarettes at the 1-month follow-up.¹⁸ A study using PRAMS data in 2016–2019 found that 46.3% of women who reported dual use of cigarettes and e-cigarettes before pregnancy stopped use during their pregnancy, while 20.2% continued dual use through the last 3 months of pregnancy, 24.9% quit e-cigarettes only, and the remaining 8.6% quit cigarette only.¹⁹

The convenience of accessing and using e-cigarettes has also been assessed. One study examined retail sales of e-cigarettes from 2017 to 2022 in 48 US states and included data from different types of stores, but excluded online or vape shop purchases of e-cigarette products.²⁰ This study concluded that there was a 293.6% increase in total monthly unit sales of e-cigarettes from 2017 to 2022.²⁰ Another study examined the reasons behind a rise in popularity in using e-cigarettes and found that some US adults reported using e-cigarettes due to their convenience.²¹ The conveniences of using e-cigarettes over traditional cigarettes identified in this article included the following: using them in some non-smoking areas, not being weather-dependent, and not needing additional supplies like lighters.²¹

Previous research among women around the time of pregnancy has primarily focused on patterns, trends, and characteristics of cigarette smoking cessation. In these studies, there has been a variation in the prevalence of quitting by multiple sociodemographic and pregnancy characteristics. Some of these studies involved US-based data,^10,11,22 whereas others included population-based data from Canada,²³ Europe,²⁴ and Australia.²⁵ Based on these studies, certain groups of women appeared more likely to quit cigarette use during pregnancy, such as those who were >20 years of age,²³ had higher education,^{10,11,22–24} had an annual income greater than $15,000,¹¹ were Hispanic or Asian/Pacific Islander,¹¹ were first-time mothers,^11,24,25 had an intended pregnancy,^11,24 were lighter smokers,^11,22 and did not use alcohol during pregnancy.^11,24 However, to our knowledge, no study to date has examined sociodemographic and pregnancy predictors of quitting among women who used both cigarettes and e-cigarettes or e-cigarettes exclusively around the time of pregnancy.

In addition, although studies have found that certain characteristics may be indicative of an individual's likelihood to use cigarettes or e-cigarettes,¹³ predictors of successfully quitting dual use have yet to be identified. Herein, we aimed to (1) examine the changes in e-cigarette use around the time of pregnancy, using the most recent US-based data, and (2) identify significant predictors of quitting cigarette use, e-cigarette use, and both during pregnancy.

Materials and Methods

Study population and design

We analyzed secondary data from PRAMS, an ongoing US national population-based surveillance project run by the Centers for Disease Control and Prevention (CDC) since 1988.¹² In PRAMS, postpartum mothers who recently had a live birth were selected from birth certificates of the participating states using stratified systematic sampling. The annual sample size per site varied from 1,000 to 3,000 and subpopulations, including mothers who had low-birth-weight infants, were oversampled.¹² Eligible mothers were first contacted through mail within 2–4 months postpartum. Mothers who did not respond were then contacted by phone. Using a survey questionnaire, mothers reported their attitudes and experiences before, during, and shortly after pregnancy.

We focused on the PRAMS data collected in the most recent Phase 8 survey from 2016 to 2019, since core questionnaire questions on e-cigarette use were initially introduced in 2016. A total of 153,336 mothers were enrolled in the Phase 8 survey, of which, 149,648 completed data on e-cigarette and cigarette use before and during late pregnancy. Our sample of women included 4,006 who reported dual use and 1,685 who reported exclusive e-cigarette use, both before pregnancy. All participating, eligible states in the Phase 8 survey were included in this analysis, regardless of differences in response rate.

PRAMS assessed the frequency of use of e-cigarettes and cigarettes differently. The survey presumed that most individuals who smoked cigarettes used them daily and captured the nuance of heaviness of use through the number of cigarettes per day, with no measure of the number of days of smoking per week. In contrast, for e-cigarettes, the survey presumed that a substantial proportion of individuals who used e-cigarettes used them less than daily, and thus there was no measure on the heaviness of daily use, but instead detailed frequency, including weekly or monthly use. These assessment differences in the frequency of use between cigarettes and e-cigarettes may be explained by differences in the dependence on these products. According to previous literature, most adults have been associated with a greater dependence on cigarettes than e-cigarettes,^26,27 and thus may smoke more often than vape.

e-Cigarette/cigarette use

Mothers reported about their e-cigarette and cigarette use separately. For e-cigarette use, mothers were asked, “On average, how often did you use e-cigarettes or other electronic nicotine products?” during the 3 months before pregnancy and during the last 3 months of pregnancy (late pregnancy).

For clarification, the survey questionnaire defined “E-cigarettes (electronic cigarettes) and other electronic nicotine vaping products (such as vape pens, e-hookahs, hookah pens, e-cigars, e-pipes) are battery-powered devices that use nicotine liquid rather than tobacco leaves and produce vapor instead of smoke.” Original responses included “more than once a day,” “once a day,” “2–6 days a week,” “1 day a week or less,” and “I did not use e-cigarettes or other electronic nicotine products then.” In the final analysis, some categories with smaller sample sizes were collapsed into one category to achieve sufficient statistical power. In the State of Hawaii only, mothers were asked about their postpartum use of e-cigarettes, “how often do you use e-cigarettes or other electronic nicotine products in an average week now?” with the same responses offered.

For cigarette use, mothers were asked the question, “how many cigarettes did you smoke on an average day?” before, during, and after pregnancy. Response options included “41 cigarettes or more,” “21 to 40 cigarettes,” “11 to 20 cigarettes,” “6 to 10 cigarettes,” “1 to 5 cigarettes,” “less than 1 cigarette,” and “I didn't smoke then.” Similar to e-cigarette use, some categories with smaller sample sizes were collapsed into one category. Postpartum cigarette use was assessed in the core questionnaire for all PRAMS participating sites.

Based on the original variables assessing e-cigarette and cigarette use in the questionnaire, we combined both types of use during the 3 months before pregnancy to define the following two different types of use: (1) no cigarette use, but yes to e-cigarette use (exclusive e-cigarette use), and (2) yes to both cigarette and e-cigarette use (dual use). The same classification was used to describe exclusive e-cigarette and dual use during the last 3 months of pregnancy and postpartum if available.

Within the dual-use group, four categories were used to describe changes in use during pregnancy: continuous dual use, quitting both cigarette and e-cigarette use, quitting cigarette use only, and quitting e-cigarette use only. To increase sample sizes per category within the exclusive e-cigarette use group, original categories of frequency (Supplementary Table S1) were collapsed from “once a day,” “2–6 days a week,” and “1 day a week or less” into “once a day or less” as one category, “more than once a day” as a second category, and “I did not use e-cigarettes or other electronic nicotine products then” as the third category, “none.”

Accordingly, we defined six dynamic changes: (1) “once a day or less” before and “none” during late pregnancy, (2) “once a day or less” before and during late pregnancy, (3) “once a day or less” before and “more than once a day” during late pregnancy, (4) “more than once a day” before and “none” during late pregnancy, (5) “more than once a day” before and “once a day or less” during late pregnancy, and (6) “more than once a day” before and during late pregnancy.

Quitting e-cigarette use during pregnancy was dichotomized as “continuous e-cigarette use” and “quitting e-cigarette use.” The same applied for quitting cigarette use or quitting both. The definition of quitting in this study referred to mothers who used e-cigarettes/cigarettes during the 3 months before pregnancy, but did not use during the last 3 months of pregnancy. This is presumably based on mothers reporting the following responses in the survey: “I didn't smoke then” or “I did not use e-cigarettes or other electronic nicotine products then.”

Predictors of dual use and exclusive e-cigarette use

Based on previous literature^13,15,28,29 and knowledge, the following characteristics were selected as potential predictors: maternal age, race,³⁰ Hispanic ethnicity, years of education, alcohol consumption in the 3 months before pregnancy, marital status, diabetes status, annual household income, health insurance type, number of prenatal care visits, infant sex, and number of previous live births. Prepregnancy body mass index (BMI) was previously calculated as weight in kilograms divided by the square of height in meters and categorized as underweight (BMI <18.5), normal weight (18.5–24.9), overweight (25–29.9), and obese (≥30).³¹ We also individually assessed the frequencies of e-cigarette use and cigarette use (among mothers who reported dual use) during the 3 months before pregnancy and hookah use around the time of pregnancy.

Statistical analysis

In the descriptive analysis, we calculated frequencies and percentages for categorical characteristics and mean and standard error (SE) estimates for continuous characteristics (Table 1). National prevalence estimates of mothers who used cigarettes and/or e-cigarettes before pregnancy, during pregnancy, or 2–6 months postpartum (Fig. 1) were also calculated. An additional descriptive figure displaying complete data of mothers who used cigarettes and/or e-cigarettes before pregnancy, during pregnancy, or 2–6 months postpartum was only reported in the state of Hawaii (Supplementary Fig. S1).

Table 1.

Sociodemographic and Pregnancy Characteristics of the Analytic Sample

Characteristica	Among mothers who exclusively used e-cigarettes before pregnancy (n = 1,685)		Among mothers who used both e-cigarettes and cigarettes before pregnancy (n = 4,006)
	Unweighted, n (%)/mean ± SE	Weighted, % (95% CI)/mean ± SEb	Unweighted, n (%)/mean ± SE	Weighted, % (95% CI)/mean ± SEb
Age, years
≤24	691 (41.0)	38.9 (38.6–39.2)	1,650 (41.2)	42.5 (42.3–42.7)
25–29	465 (27.6)	28.9 (28.7–29.2)	1,192 (29.8)	28.6 (28.4–28.8)
30–34	342 (20.3)	21.3 (21.0–21.6)	790 (19.7)	19.4 (19.2–19.5)
≥35	187 (11.1)	10.9 (10.7–11.1)	372 (9.3)	9.6 (9.5–9.7)
Race
White	1,144 (69.4)	79.1 (78.8–79.4)	2,739 (70.9)	84.9 (84.8–85.1)
African American	196 (11.9)	9.4 (9.2–9.6)	405 (10.5)	6.9 (6.8–7.0)
American Indian or Alaska Native	62 (3.8)	1.0 (0.9–1.0)	236 (6.1)	1.2 (1.1–1.2)
Asian	38 (2.3)	1.9 (1.8–2.0)	74 (1.9)	1.3 (1.3–1.4)
Other non-White/Mixed race	209 (12.7)	8.7 (8.5–8.9)	411 (10.6)	5.6 (5.5–5.7)
Ethnicity
Non-Hispanic	1,411 (84.7)	88.5 (88.3–88.7)	3,410 (88.1)	91.3 (91.2–91.5)
Hispanic	254 (15.3)	11.5 (11.3–11.7)	459 (11.9)	8.7 (8.6–8.8)
Prepregnancy BMI, kg/m2, Mean (±SE)	27.1 ± 0.2	27.0 ± 0.2	26.9 ± 0.1	26.8 ± 0.1
Underweight (<18.5)	66 (4.0)	3.8 (3.7–3.9)	207 (5.3)	4.6 (4.5–4.7)
Normal (18.5–24.9)	687 (41.6)	43.0 (42.6–43.3)	1,583 (40.5)	40.5 (40.3–40.7)
Overweight (25.0–29.9)	403 (24.4)	24.1 (23.9–24.4)	972 (24.9)	26.3 (26.1–26.5)
Obese (≥30.0)	494 (29.9)	29.1 (28.8–29.4)	1,149 (29.4)	28.5 (28.3–28.7)
Alcohol consumption during 3 months before pregnancy
Did not drink then	177 (12.5)	12.4 (12.1–12.6)	490 (14.5)	15.0 (14.8–15.2)
<1 drink a week	490 (34.7)	34.4 (34.1–34.7)	1,135 (33.5)	32.8 (32.6–33.0)
≥1 drink a week	747 (52.8)	53.3 (52.9–53.6)	1,766 (52.1)	52.2 (52.0–52.5)
Education, years
≤12	702 (42.2)	42.8 (42.5–43.2)	2,285 (57.4)	56.5 (56.3–56.7)
13–15	666 (40.0)	36.7 (36.4–37.1)	1,370 (34.4)	34.4 (34.2–34.6)
≥16	296 (17.8)	20.4 (20.2–20.7)	323 (8.1)	9.1 (9.0–9.3)
Marital status
Unmarried	911 (54.2)	53.4 (53.0–53.7)	2,750 (68.9)	67.6 (67.4–67.8)
Married	771 (45.8)	46.6 (46.3–47.0)	1,242 (31.1)	32.4 (32.2–32.6)
Diabetes
No	1,567 (93.1)	93.3 (93.1–93.4)	3,750 (93.8)	94.1 (94.0–94.2)
Yes	116 (6.9)	6.7 (6.6–6.9)	247 (6.2)	5.9 (5.8–6.0)
Annual household income, US dollars
≤ $16,000	341 (23.1)	22.8 (22.5–23.1)	1,371 (38.7)	36.9 (36.7–37.2)
$16,001 to $32,000	440 (29.7)	28.2 (27.9–28.5)	1,185 (33.4)	31.7 (31.5–31.9)
$32,001 to $85,000	505 (34.1)	34.3 (33.9–34.6)	745 (21.0)	22.2 (22.1–22.4)
≥ $85,001	193 (13.0)	14.8 (14.6–15.1)	242 (6.8)	9.1 (9.0–9.3)
Health insurance
Medicaid/governmental	860 (51.6)	47.7 (47.4–48.1)	2,742 (69.0)	65.5 (65.3–65.7)
Nongovernmental	806 (48.4)	52.3 (51.9–52.6)	1,230 (31.0)	34.6 (34.3–34.8)
Number of prenatal care visits
≤8	328 (20.1)	15.4 (15.2–15.6)	1,066 (27.4)	21.9 (21.7–22.1)
9–11	522 (31.9)	32.5 (32.2–32.9)	1,157 (29.8)	29.4 (29.2–29.6)
≥12	785 (48.0)	52.1 (51.7–52.4)	1,662 (42.8)	48.7 (48.5–48.9)
Infant sex
Male	824 (48.9)	47.7 (47.4–48.0)	2,032 (50.7)	53.6 (53.3–53.8)
Female	861 (51.1)	52.3 (52.0–52.7)	1,973 (49.3)	46.5 (46.2–46.7)
Number of previous live births
0	914 (54.4)	55.3 (55.0–55.7)	1,813 (45.4)	48.2 (48.0–48.4)
≥1	767 (45.6)	44.7 (44.3–45.0)	2,182 (54.6)	51.8 (51.6–52.0)
Frequency of e-cigarette use during the 3 months before pregnancy
1 day a week or less	532 (31.6)	29.9 (29.7–30.2)	1,855 (46.3)	46.1 (45.9–46.3)
2–6 days a week	181 (10.7)	11.2 (11.0–11.4)	607 (15.2)	14.1 (14.0–14.3)
Once a day	173 (10.3)	10.4 (10.2–10.6)	376 (9.4)	9.0 (8.9–9.1)
More than once a day	799 (47.4)	48.5 (48.2–48.8)	1,168 (29.2)	30.7 (30.5–30.9)
Frequency of e-cigarette use during the 3 months before pregnancy
Once a day or less	886 (52.6)	51.5 (51.2–51.8)	2,838 (70.8)	69.3 (69.1–69.5)
More than once a day	799 (47.4)	48.5 (48.2–48.8)	1,168 (29.2)	30.7 (30.5–30.9)
Frequency of cigarette use during the 3 months before pregnancy
<1 cigarette a day	N/A	N/A	349 (8.7)	10.1 (10.0–10.2)
1–5 cigarettes a day	N/A	N/A	1,139 (28.4)	25.9 (25.7–26.1)
6–10 cigarettes a day	N/A	N/A	1,102 (27.5)	27.3 (27.1–27.5)
11–20 cigarettes a day	N/A	N/A	1,022 (25.5)	27.4 (27.2–27.6)
21–40 cigarettes a day	N/A	N/A	307 (7.7)	7.2 (7.1–7.3)
41 or more cigarettes a day	N/A	N/A	87 (2.2)	2.1 (2.0–2.1)
Frequency of cigarette use during the 3 months before pregnancy
≤5 cigarettes a day	N/A	N/A	1,488 (37.1)	36.0 (35.8–36.2)
6–10 cigarettes a day	N/A	N/A	1,102 (27.5)	27.3 (27.1–27.5)
≥11 cigarettes a day	N/A	N/A	1,416 (35.3)	36.7 (36.5–36.9)
Hookah use around the time of pregnancyc
No	1,144 (76.8)	75.8 (75.5–76.1)	2,994 (78.9)	78.7 (78.5–78.9)
Yes	345 (23.2)	24.2 (23.9–24.5)	799 (21.1)	21.3 (21.1–21.5)

^a Sum of categories may not be equal to the total sample size due to missing data on some characteristics.

^b Results were weighted to be representative of each state's population by accounting for nonresponse, noncoverage, and non-random sampling.

^c Within the 2 years prior to the 2- to 6-month postpartum survey, including before, during, and shortly after pregnancy.

BMI, body mass index; CI, confidence interval; e-cigarettes, electronic cigarettes; SE, standard error.

FIG. 1.

National prevalence of mothers who used e-cigarettes, cigarettes, or both e-cigarettes and cigarettes before, during, and after* pregnancy, 2016–2019. *The postpartum data are only available for cigarette use. e-cigarettes, electronic cigarettes.

Dynamic changes of exclusive e-cigarette use during late pregnancy were displayed using frequencies (Fig. 2). Further description of the distribution of original frequency categories (“1 day a week or less” and “2-6 days a week”) of use before pregnancy binned into the “less than once a day” category in this figure is also provided (Supplementary Table S1). Odds ratios (OR) with 95% confidence intervals (CI) from multinomial and binomial logistic regression models were used to identify significant predictors for quitting use among mothers who reported exclusive e-cigarette use (Table 2) or dual use (Table 3), respectively. The multiple imputation procedure was used to handle missing data on predictors.³²

FIG. 2.

Dynamic change in frequency of using e-cigarettes from before to late pregnancy among mothers who exclusively used e-cigarettes before pregnancy, 2016–2019.

Table 2.

Sociodemographic and Pregnancy Predictors of Quitting During Late Pregnancy Among Mothers Who Exclusively Used Electronic Cigarettes Before Pregnancy (N = 1,685)

Characteristic	Quitting e-cigarette use during late pregnancya
	%	OR (95% CI)	p
Among the total sample	81.5
Age, years
≤24	85.6	Reference
25–29	83.7	0.86 (0.82–0.90)	<0.001
30–34	72.5	0.44 (0.42–0.46)	<0.001
≥35	78.5	0.61 (0.58–0.65)	<0.001
Race
White	81.5	Reference
African American	85.0	1.29 (1.19–1.40)	<0.001
American Indian or Alaska Native	81.8	1.02 (0.81–1.29)	0.849
Asian	91.2	2.32 (1.92–2.82)	<0.001
Other non-White/Mixed race	75.2	0.70 (0.64–0.76)	<0.001
Ethnicity
Non-Hispanic	81.7	Reference
Hispanic	80.0	0.90 (0.85–0.95)	<0.001
Prepregnancy BMI, kg/m2		1.02 (1.01–1.02)	<0.001
Underweight (<18.5)	70.3	0.46 (0.42–0.51)	<0.001
Normal (18.5–24.9)	84.0	Reference
Overweight (25.0–29.9)	78.7	0.72 (0.67–0.77)	<0.001
Obese (≥30.0)	81.7	0.88 (0.84–0.92)	<0.001
Alcohol consumption during 3 months before pregnancy
Did not drink then	80.5	Reference
<1 drink a week	86.8	1.31 (0.95–1.81)	0.099
≥1 drink a week	85.4	1.25 (0.91–1.72)	0.166
Education, years
≤12	76.3	Reference
13–15	85.2	1.78 (1.71–1.86)	<0.001
≥16	85.6	1.84 (1.74–1.94)	<0.001
Marital status
Unmarried	80.6	Reference
Married	82.6	1.14 (1.10–1.18)	<0.001
Diabetes
No	81.7	Reference
Yes	78.1	0.80 (0.75–0.85)	<0.001
Annual household income, US dollars
≤ $16,000	74.6	Reference
$16,001 to $32,000	78.7	1.23 (1.08–1.40)	0.004
$32,001 to $85,000	86.4	2.02 (1.76–2.32)	<0.001
≥ $85,001	83.3	1.65 (1.50–1.81)	<0.001
Health insurance
Medicaid/governmental	76.0	Reference
Nongovernmental	86.7	2.04 (1.91–2.18)	<0.001
Number of prenatal care visits
≤8	78.0	Reference
9–11	80.9	1.20 (1.09–1.31)	<0.001
≥12	83.0	1.37 (1.24–1.52)	<0.001
Infant sex
Male	79.8	Reference
Female	83.0	1.24 (1.19–1.28)	<0.001
Number of previous live births
0	85.2	Reference
≥1	76.9	0.58 (0.56–0.60)	<0.001
Frequency of e-cigarette use during the 3 months before pregnancy
Once a day or less	88.2	Reference
More than once a day	74.4	0.39 (0.38–0.40)	<0.001
Hookah use around the time of pregnancyb
No	80.7	Reference
Yes	86.9	1.50 (1.20–1.86)	0.001

^a Results were weighted to be representative of each state's population by accounting for nonresponse, noncoverage, and non-random sampling.

^b Within the 2 years prior to the 2- to 6-month postpartum survey, including before, during, and shortly after pregnancy.

BMI, body mass index; CI, confidence interval; e-cigarettes, electronic cigarettes; OR, odds ratio.

Table 3.

Sociodemographic and Pregnancy Predictors of Quitting During Late Pregnancy Among Mothers Who Used Both Electronic Cigarettes and Cigarettes Before Pregnancy (N = 4,006)

Characteristic	Continuous use (reference),a %	Quitting both vs. continuous usea			Quitting cigarette use only vs. continuous usea			Quitting e-cigarette use only vs. continuous usea
		%	OR (95% CI)	p	%	OR (95% CI)	p	%	OR (95% CI)	p
Among the total sample	20.7	46.2			6.6			26.5
Age, years
≤24	17.4	48.4	Reference		7.8	Reference		26.4	Reference
25–29	23.7	44.2	0.67 (0.65–0.69)	<0.001	5.4	0.51 (0.48–0.55)	<0.001	26.6	0.74 (0.72–0.76)	<0.001
30–34	25.7	44.1	0.62 (0.60–0.64)	<0.001	5.3	0.47 (0.43–0.51)	<0.001	25.0	0.64 (0.62–0.66)	<0.001
≥35	17.0	46.5	0.98 (0.94–1.03)	0.418	6.5	0.85 (0.77–0.94)	0.003	30.1	1.16 (1.11–1.22)	<0.001
Race
White	21.3	45.4	Reference		6.7	Reference		26.6	Reference
African American	17.4	43.9	1.18 (1.09–1.27)	<0.001	6.6	1.19 (1.02–1.40)	0.032	32.1	1.46 (1.36–1.57)	<0.001
American Indian or Alaska Native	27.0	41.1	0.70 (0.59–0.82)	<0.001	3.0	0.46 (0.14–1.58)	0.207	28.8	0.85 (0.69–1.05)	0.128
Asian	12.6	73.8	2.72 (2.41–3.08)	<0.001	5.3	1.32 (1.06–1.66)	0.014	8.2	0.52 (0.43–0.62)	<0.001
Other non-White/Mixed race	16.6	55.9	1.56 (1.40–1.73)	<0.001	3.5	0.68 (0.48–0.95)	0.027	23.9	1.14 (1.02–1.26)	0.020
Ethnicity
Non-Hispanic	21.4	44.4	Reference		6.4	Reference		27.7	Reference
Hispanic	12.5	65.5	2.49 (2.28–2.72)	<0.001	7.6	2.03 (1.72–2.40)	<0.001	14.4	0.89 (0.80–0.99)	0.031
Prepregnancy BMI, kg/m2			1.02 (1.02–1.02)	<0.001		1.02 (1.01–1.02)	<0.001		1.03 (1.02–1.03)	<0.001
Underweight (<18.5)	29.9	36.0	0.56 (0.49–0.63)	<0.001	1.7	0.19 (0.15–0.22)	<0.001	32.4	0.84 (0.73–0.97)	0.017
Normal (18.5–24.9)	20.8	45.8	Reference		6.5	Reference		27.0	Reference
Overweight (25.0–29.9)	20.9	48.3	1.07 (1.01–1.14)	0.033	8.2	1.28 (1.20–1.37)	<0.001	22.5	0.85 (0.80–0.91)	<0.001
Obese (≥30.0)	17.8	47.6	1.23 (1.17–1.29)	<0.001	6.5	1.19 (1.12–1.26)	<0.001	28.0	1.25 (1.18–1.32)	<0.001
Alcohol consumption during 3 months before pregnancy
Did not drink then	29.6	29.7	Reference		12.9	Reference		27.8	Reference
<1 drink a week	23.9	44.7	1.72 (1.48–1.99)	<0.001	5.5	0.53 (0.42–0.69)	<0.001	25.8	1.11 (0.94–1.30)	0.213
≥1 drink a week	12.3	58.1	3.73 (3.21–4.33)	<0.001	5.4	0.93 (0.72–1.20)	0.565	24.1	1.77 (1.54–2.03)	<0.001
Education, years
≤12	25.4	39.1	Reference		5.8	Reference		29.7	Reference
13–15	17.1	49.0	1.86 (1.80–1.93)	<0.001	7.4	1.88 (1.76–2.01)	<0.001	26.6	1.33 (1.29–1.38)	<0.001
≥16	4.8	79.9	10.53 (8.99–12.34)	<0.001	7.5	6.58 (5.63–7.69)	<0.001	7.7	1.34 (1.14–1.58)	0.001
Marital status
Unmarried	22.2	42.6	Reference		6.9	Reference		28.3	Reference
Married	17.8	53.8	1.58 (1.54–1.62)	<0.001	5.7	1.04 (0.98–1.10)	0.210	22.6	1.00 (0.97–1.03)	0.940
Diabetes
No	21.2	45.5	Reference		6.6	Reference		26.7	Reference
Yes	13.8	55.9	1.87 (1.77–1.99)	<0.001	5.3	1.22 (1.11–1.34)	<0.001	25.0	1.43 (1.34–1.53)	<0.001
Annual household income, US dollars
≤ $16,000	28.9	32.1	Reference		7.1	Reference		31.9	Reference
$16,001 to $32,000	21.1	45.9	1.86 (1.68–2.05)	<0.001	4.6	0.91 (0.77–1.08)	0.257	28.5	1.20 (1.10–1.31)	<0.001
$32,001 to $85,000	14.2	56.3	3.25 (2.93–3.60)	<0.001	8.0	2.14 (1.93–2.39)	<0.001	21.5	1.35 (1.21–1.51)	<0.001
≥ $85,001	5.8	77.5	9.94 (7.70–12.83)	<0.001	6.2	3.81 (2.87–5.05)	<0.001	10.5	1.57 (1.16–2.11)	0.005
Health insurance
Medicaid/governmental	25.7	38.5	Reference		6.3	Reference		29.6	Reference
Nongovernmental	11.1	61.4	3.64 (3.45–3.83)	<0.001	7.0	2.54 (2.34–2.77)	<0.001	20.5	1.58 (1.49–1.68)	<0.001
Number of prenatal care visits
≤8	28.8	33.2	Reference		5.4	Reference		32.5	Reference
9–11	20.6	46.2	1.89 (1.79–1.98)	<0.001	7.1	1.82 (1.61–2.05)	<0.001	26.1	1.12 (1.04–1.21)	0.005
≥12	17.8	51.6	2.42 (2.30–2.55)	<0.001	6.5	1.91 (1.69–2.17)	<0.001	24.1	1.19 (1.12–1.27)	<0.001
Infant sex
Male	19.9	45.3	Reference		7.1	Reference		27.6	Reference
Female	21.7	47.1	0.96 (0.93–0.98)	<0.001	5.9	0.76 (0.73–0.79)	<0.001	25.3	0.84 (0.82–0.86)	<0.001
Number of previous live births
0	15.6	56.3	Reference		6.9	Reference		21.1	Reference
≥1	25.6	36.9	0.40 (0.39–0.41)	<0.001	6.2	0.54 (0.52–0.57)	<0.001	31.3	0.91 (0.88–0.94)	<0.001
Frequency of e-cigarette use during the 3 months before pregnancy
Once a day or less	22.1	44.3	Reference		4.9	Reference		28.6	Reference
More than once a day	17.7	50.3	1.42 (1.38–1.45)	<0.001	10.2	2.58 (2.47–2.68)	<0.001	21.8	0.95 (0.92–0.98)	0.001
Frequency of cigarette use during the 3 months before pregnancy
≤5 cigarettes a day	7.7	72.5	Reference		7.3	Reference		12.5	Reference
6–10 cigarettes a day	22.3	39.6	0.19 (0.18–0.20)	<0.001	8.7	0.41 (0.39–0.43)	<0.001	29.4	0.82 (0.78–0.85)	<0.001
≥11 cigarettes a day	32.4	25.1	0.08 (0.08–0.09)	<0.001	4.2	0.14 (0.13–0.15)	<0.001	38.2	0.73 (0.70–0.76)	<0.001
Hookah use around the time of pregnancyb
No	23.3	42.0	Reference		5.6	Reference		29.0	Reference
Yes	11.8	60.6	2.68 (2.41–2.98)	<0.001	6.9	2.22 (1.77–2.78)	<0.001	20.7	1.38 (1.25–1.52)	<0.001

^a Results were weighted to be representative of each state's population by accounting for nonresponse, noncoverage, and non-random sampling.

^b Within the 2 years prior to the 2- to 6-month postpartum survey, including before, during, and shortly after pregnancy.

BMI, body mass index; CI, confidence interval; e-cigarettes, electronic cigarettes; OR, odds ratio.

We applied sample weights in all analyses to make our results representative of the whole US population by accounting for potential biases due to nonresponse, noncoverage, and non-random sampling.¹² All analyses in this study were conducted using SAS 9.4 software (SAS Institute, Inc., Cary, NC) and a two-sided p < 0.05 was considered statistically significant. This analysis was not preregistered and all results should be considered exploratory.

Results

Sample characteristics

As shown in Table 1, among mothers who reported dual use (n = 4,006), 71.1% (weighted) of the mothers were younger than 30 years of age. Approximately 84.9% were White, 6.9% were African Americans, 8.7% were Hispanic, 56.5% had a high school education or less, 32.4% were married, 40.5% had normal prepregnancy BMI, while 26.3% and 28.5% were overweight and obese, respectively, 52.2% had higher alcohol consumption before pregnancy, 48.7% completed 12 or more prenatal care visits, and 51.8% had at least one previous live birth. Among mothers who exclusively used e-cigarettes (n = 1,685), 67.8% of the mothers were younger than 30 years of age. Around 79.1% were White, 9.4% were African Americans, 11.5% were Hispanic, and 57.1% had a college degree or higher.

Trajectories of e-cigarette and cigarette use around pregnancy

The national prevalence of e-cigarette use, cigarette use, and dual use decreased significantly (p < 0.001) between before and during late pregnancy (Fig. 1). During both time periods, cigarette use was the most popular followed by dual use and e-cigarette use. The prevalence of all three types of use in Hawaii was similar to the national trends. During the 2- to 6-month postpartum period, a considerable proportion of mothers returned to e-cigarette and/or cigarette use (p < 0.001), which remained below the prepregnancy level (Supplementary Fig. S1).

Among mothers who vaped for once a day or less during the 3 months before pregnancy, 88.2% quit altogether, 11.1% stayed in the same category of frequency, and 0.7% increased to more than once a day during late pregnancy (Fig. 2). Among mothers who vaped for more than once a day during the 3 months before pregnancy, 74.4% quit altogether, 20.4% stayed in the same category of frequency, and 5.2% decreased to once a day or less during late pregnancy. Furthermore, in an additional in-depth analysis, regardless of the frequency of use before pregnancy among women who exclusively used e-cigarettes, patterns of use during late pregnancy were similar, in which cessation rates were the highest, followed by mothers who stayed in the same category of frequency of use, and mothers who decreased or increased use, respectively (Supplementary Table S1).

Predictors for quitting among mothers who reported exclusive e-cigarette use

Compared to mothers who were 24 years of age or younger, mothers 30–34 years of age were least likely to quit e-cigarette use during late pregnancy (72.5% vs. 85.6%; OR: 0.44 [95% CI: 0.42–0.46]); mothers 25–29 years of age (83.7% vs. 85.6%; 0.86 [0.82–0.90]) and 35 years of age or older (78.5% vs. 85.6%; 0.61 [0.58–0.65]) were also less likely to quit (Table 2). Compared to White mothers, Asian mothers were most likely to quit (91.2% vs. 81.5%; 2.32 [1.92–2.82]) e-cigarette use. In addition, African American mothers (85.0% vs. 81.5%; 1.29 [1.19–1.40]) had a higher likelihood to quit, while mothers who were other non-White or of mixed race (75.2% vs. 81.5%; 0.70 [0.64–0.76]) were least likely to quit.

Compared to mothers with a normal prepregnancy BMI, mothers who were underweight (70.3% vs. 84.0%; 0.46 [0.42–0.51]) were least likely to quit. Furthermore, mothers who were overweight (78.7% vs. 84.0%; 0.72 [0.67–0.77]) or obese (81.7% vs. 84.0%; 0.88 [0.84–0.92]) also had a lower likelihood to quit.

Compared to non-Hispanic mothers, Hispanic mothers had a lower likelihood to quit e-cigarette use (80.0% vs. 81.7%; 0.90 [0.85–0.95]). Compared to mothers with ≤12 years of education, mothers with ≥16 years of education (85.6% vs. 76.3%; 1.84 [1.74–1.94]) had the highest likelihood to quit e-cigarette use; mothers with 13–15 years of education also had a higher likelihood to quit (85.2% vs. 76.3%; 1.78 [1.71–1.86]). Compared to unmarried mothers, married mothers had a higher likelihood to quit e-cigarettes (82.6% vs. 80.6%; 1.14 [1.10–1.18]). In contrast, compared to mothers without diabetes, mothers with diabetes had a lower likelihood to quit (78.1% vs. 81.7%; 0.80 [0.75–0.85]).

Compared to mothers with an annual household income of ≤ $16,000, those with a greater income had a higher likelihood to quit e-cigarette use, including mothers with an annual household income of $16,001 to $32,000 (78.7% vs. 74.6%; 1.23 [1.08–1.40]), $32,001 to $85,000 (86.4% vs. 74.6%; 2.02 [1.76–2.32]), and $85,001 or more (83.3% vs. 74.6%; 1.65 [1.50–1.81]). Compared to mothers with Medicaid or other governmental health insurance, those with nongovernmental health insurance had a higher likelihood to quit e-cigarette use (86.7% vs. 76.0%; 2.04 [1.91–2.18]). Compared to mothers with ≤8 prenatal care visits, those with a higher number of visits had a higher likelihood to quit e-cigarette use, including 9–11 visits (80.9% vs. 78.0%; 1.20 [1.09–1.31]) and ≥12 visits (83.0% vs. 78.0%; 1.37 [1.24–1.52]).

Compared to mothers who gave birth to male infants, mothers who had female infants had a higher likelihood to quit e-cigarette use (83.0% vs. 79.8%; 1.24 [1.19–1.28]). Compared to mothers with no previous live birth, mothers who had one or more births had a lower likelihood to quit e-cigarette use (76.9% vs. 85.2%; 0.58 [0.56–0.60]). Compared to mothers who used e-cigarettes once a day or less during the 3 months before pregnancy, mothers who used more than once a day had a lower likelihood to quit e-cigarette use (74.4% vs. 88.2%; 0.39 [0.38–0.40]). Compared to mothers who did not use hookah around the time of pregnancy, including before and during pregnancy, mothers who used hookah had a greater likelihood to quit e-cigarettes (86.9% vs. 80.7%; 1.50 [1.20–1.86]).

Predictors for quitting among mothers who reported dual use

Compared to mothers 24 years of age or younger, mothers 30–34 years of age were least likely to quit both e-cigarette and cigarette use during late pregnancy (44.1% vs. 48.4%; OR: 0.62 [95% CI: 0.60–0.64]); mothers 25–29 years of age (44.2% vs. 48.4%; 0.67 [0.65–0.69]) were also less likely to quit both (continuous dual use as the reference outcome category) (Table 3). Compared to White mothers, Asian mothers had the highest likelihood of quitting both (73.8% vs. 45.4%; 2.72 [2.41–3.08]); African American (43.9% vs. 45.4%; 1.18 [1.09–1.27]) and other non-White or mixed race (55.9% vs. 45.4%; 1.56 [1.40–1.73]) mothers also had a higher likelihood to quit both, whereas American Indian or Alaska Native mothers had a lower likelihood of quitting both (41.1% vs. 45.4%; 0.70 [0.59–0.82]).

Compared to non-Hispanic mothers, Hispanic mothers had a higher likelihood of quitting both (65.5% vs. 44.4%; 2.49 [2.28–2.72]). Compared to mothers with a normal prepregnancy BMI, mothers who were underweight had a lower likelihood of quitting both (36.0% vs. 45.8%; 0.56 [0.49–0.63]), while mothers who were overweight (48.3% vs. 45.8%; 1.07 [1.01–1.14]) or obese (47.6% vs. 45.8%; 1.23 [1.17–1.29]) had a higher likelihood of quitting both. Compared to mothers who did not consume alcohol 3 months before pregnancy, mothers who had <1 drink a week (44.7% vs. 29.7%; 1.72 [1.48–1.99]) or ≥1 drink a week (58.1% vs. 29.7%; 3.73 [3.21–4.33]) had a higher likelihood of quitting both.

Compared to mothers with ≤12 years of education, mothers who had 13–15 years (49.0% vs. 39.1%; 1.86 [1.80–1.93]) or ≥16 years of education (79.9% vs. 39.1%; 10.53 [8.99–12.34]) had a higher likelihood of quitting both. Compared to unmarried mothers, married mothers had a higher likelihood of quitting both (53.8% vs. 42.6%; 1.58 [1.54–1.62]). Compared to mothers without diabetes, mothers with diabetes had a higher likelihood of quitting both (55.9% vs. 45.5%; 1.87 [1.77–1.99]). Compared to mothers with an annual household income of ≤ $16,000, those with a greater income had a higher likelihood to quit both, including mothers with an annual household income of $16,001 to $32,000 (45.9% vs. 32.1%; 1.86 [1.68–2.05]), $32,001 to $85,000 (56.3% vs. 32.1%; 3.25 [2.93–3.60]), and $85,001 or more (77.5% vs. 32.1%; 9.94 [7.70–12.83]).

Compared to mothers with Medicaid or governmental health insurance, those with nongovernmental health insurance had a higher likelihood to quit both (61.4% vs. 38.5%; 3.64 [3.45–3.83]). Compared to mothers with ≤8 prenatal care visits, those with a higher number of visits had a higher likelihood to quit both, including 9–11 visits (46.2% vs. 33.2%; 1.89 [1.79–1.98]) and ≥12 visits (51.6% vs. 33.2%; 2.42 [2.30–2.55]). Compared to mothers who gave birth to male infants, mothers who had female infants had a lower likelihood to quit both (47.1% vs. 45.3%; 0.96 [0.93–0.98]). Compared to mothers with no previous live birth, mothers who had one or more births had a lower likelihood to quit both (36.9% vs. 56.3%; 0.40 [0.39–0.41]).

Compared to mothers who used e-cigarettes once a day or less during the 3 months before pregnancy, mothers who used e-cigarettes more than once a day had a higher likelihood to quit both (50.3% vs. 44.3%; 1.42 [1.38–1.45]). Compared to mothers who used ≤5 cigarettes a day during the 3 months before pregnancy, those with a greater frequency of use had a lower likelihood of quitting both, including 6–10 cigarettes a day (39.6% vs. 72.5%; 0.19 [0.18–0.20]) and ≥11 cigarettes a day (25.1% vs. 72.5%; 0.08 [0.08–0.09]). Compared to mothers who did not use hookah around the time of pregnancy, including before and during pregnancy, mothers who used hookah had a greater likelihood to quit both (60.6% vs. 42.0%; 2.68 [2.41–2.98]).

Compared to mothers 24 years of age or younger, mothers 30–34 years of age were least likely to quit cigarette use only during late pregnancy (5.3% vs. 7.8%; OR: 0.47 [95% CI: 0.43–0.51]); mothers 25–29 years of age (5.4% vs. 7.8%; 0.51 [0.48–0.55]) and 35 years of age or older (6.5% vs. 7.8%; 0.85 [0.77–0.94]) were also less likely to quit cigarette use only (continuous dual use as the reference outcome category). Compared to White mothers, African American (6.6% vs. 6.7%; 1.19 [1.02–1.40]) and Asian (5.3% vs. 6.7%; 1.32 [1.06–1.66]) mothers had a higher likelihood of quitting cigarette use only, whereas other non-White or mixed-race mothers had a lower likelihood of quitting cigarette use only (3.5% vs. 6.7%; 0.68 [0.48–0.95]). Compared to non-Hispanic mothers, Hispanic mothers had a higher likelihood of quitting cigarette use only (7.6% vs. 6.4%; 2.03 [1.72–2.40]).

Compared to mothers with a normal prepregnancy BMI, mothers who were underweight had a lower likelihood of quitting cigarette use only (1.7% vs. 6.5%; 0.19 [0.15–0.22]), while mothers who were overweight (8.2% vs. 6.5%; 1.28 [1.20–1.37]) or obese (6.5% vs. 6.5%; 1.19 [1.12–1.26]) had a higher likelihood of quitting cigarette use only. Compared to mothers who did not consume alcohol 3 months before pregnancy, mothers who had <1 drink a week (5.5% vs. 12.9%; 0.53 [0.42–0.69]) had a lower likelihood of quitting cigarette use only. Compared to mothers with ≤12 years of education, mothers who had 13–15 years (7.4% vs. 5.8%; 1.88 [1.76–2.01]) or ≥16 years of education (7.5% vs. 5.8%; 6.58 [5.63–7.69]) had a higher likelihood of quitting cigarette use only.

Compared to mothers without diabetes, mothers with diabetes had a higher likelihood of quitting cigarette use only (5.3% vs. 6.6%; 1.22 [1.11–1.34]). Compared to mothers with an annual household income of ≤ $16,000, those with a greater income had a higher likelihood to quit cigarette use only, including mothers with an annual household income of $32,001 to $85,000 (8.0% vs. 7.1%; 2.14 [1.93–2.39]) and $85,001 or more (6.2% vs. 7.1%; 3.81 [2.87–5.05]).

Compared to mothers with Medicaid or governmental health insurance, those with nongovernmental health insurance had a higher likelihood to quit cigarette use only (7.0% vs. 6.3%; 2.54 [2.34–2.77]). Compared to mothers with ≤8 prenatal care visits, those with a higher number of visits had a higher likelihood to quit cigarette use only, including 9–11 visits (7.1% vs. 5.4%; 1.82 [1.61–2.05]) and ≥12 visits (6.5% vs. 5.4%; 1.91 [1.69–2.17]). Compared to mothers who gave birth to male infants, mothers who had female infants had a lower likelihood to quit cigarette use only (5.9% vs. 7.1%; 0.76 [0.73–0.79]). Compared to mothers with no previous live birth, mothers who had one or more births had a lower likelihood to quit cigarette use only (6.2% vs. 6.9%; 0.54 [0.52–0.57]).

Compared to mothers who used e-cigarettes once a day or less during the 3 months before pregnancy, mothers who used e-cigarettes more than once a day had a higher likelihood to quit cigarette use only (10.2% vs. 4.9%; 2.58 [2.47–2.68]). Compared to mothers who used ≤5 cigarettes a day during the 3 months before pregnancy, those with a greater frequency of use had a lower likelihood of quitting cigarette use only, including 6–10 cigarettes a day (8.7% vs. 7.3%; 0.41 [0.39–0.43]) and ≥11 cigarettes a day (4.2% vs. 7.3%; 0.14 [0.13–0.15]). Compared to mothers who did not use hookah around the time of pregnancy, mothers who used hookah had a greater likelihood to quit cigarette use only (6.9% vs. 5.6%; 2.22 [1.77–2.78]).

Compared to mothers 24 years of age or younger, mothers 30–34 years of age were least likely to quit e-cigarette use only during late pregnancy (25.0% vs. 26.4%; OR: 0.64 [95% CI: 0.62–0.66]); mothers 25–29 years of age (26.6% vs. 26.4%; 0.74 [0.72–0.76]) were also less likely to quit, while mothers 35 years of age or older were more likely to quit e-cigarette use only (30.1% vs. 26.4%; 1.16 [1.11–1.22]) (continuous dual use as the reference outcome category). Compared to White mothers, Asian mothers had a lower likelihood of quitting e-cigarette use only (8.2% vs. 26.6%; 0.52 [0.43–0.62]), whereas African American (32.1% vs. 26.6%; 1.46 [1.36–1.57]) and other non-White or mixed-race (23.9% vs. 26.6%; 1.14 [1.02–1.26]) mothers had a higher likelihood to quit e-cigarette use only.

Compared to non-Hispanic mothers, Hispanic mothers had a lower likelihood of quitting e-cigarette use only (14.4% vs. 27.7%; 0.89 [0.80–0.99]). Compared to mothers with a normal prepregnancy BMI, mothers who were underweight (32.4% vs. 27.0%; 0.84 [0.73–0.97]) or overweight (22.5% vs. 27.0%; 0.85 [0.80–0.91]) had a lower likelihood of quitting e-cigarette use only, while mothers who were obese had a higher likelihood of quitting e-cigarette use only (28.0% vs. 27.0%; 1.25 [1.18–1.32]). Compared to mothers who did not consume alcohol 3 months before pregnancy, mothers who had ≥1 drink a week (24.1% vs. 27.8%; 1.77 [1.54–2.03]) had a higher likelihood of quitting e-cigarette use only.

Compared to mothers with ≤12 years of education, mothers who had 13–15 years (26.6% vs. 29.7%; 1.33 [1.29–1.38]) or ≥16 years of education (7.7% vs. 29.7%; 1.34 [1.14–1.58]) had a higher likelihood of quitting e-cigarette use only. Compared to mothers without diabetes, mothers with diabetes had a higher likelihood of quitting e-cigarette use only (25.0% vs. 26.7%; 1.43 [1.34–1.53]). Compared to mothers with an annual household income of ≤ $16,000, those with a greater income had a higher likelihood to quit e-cigarette use only, including mothers with an annual household income of $16,001 to $32,000 (28.5% vs. 31.9%; 1.20 [1.10–1.31]), $32,001 to $85,000 (21.5% vs. 31.9%; 1.35 [1.21–1.51]), and $85,001 or more (10.5% vs. 31.9%; 1.57 [1.16–2.11]).

Compared to mothers with Medicaid or governmental health insurance, those with nongovernmental health insurance had a higher likelihood to quit e-cigarette use only (20.5% vs. 29.6%; 1.58 [1.49–1.68]). Compared to mothers with ≤8 prenatal care visits, those with a higher number of visits had a higher likelihood to quit e-cigarette use only, including 9–11 visits (26.1% vs. 32.5%; 1.12 [1.04–1.21]) and ≥12 visits (24.1% vs. 32.5%; 1.19 [1.12–1.27]). Compared to mothers who gave birth to male infants, mothers who had female infants had a lower likelihood to quit e-cigarette use only (25.3% vs. 27.6%; 0.84 [0.82–0.86]). Compared to mothers with no previous live birth, mothers who had one or more births had a lower likelihood to quit e-cigarette use only (31.3% vs. 21.1%; 0.91 [0.88–0.94]).

Compared to mothers who used e-cigarettes once a day or less during the 3 months before pregnancy, mothers who used e-cigarettes more than once a day had a lower likelihood to quit e-cigarette use only (21.8% vs. 28.6%; 0.95 [0.92–0.98]). Compared to mothers who used ≤5 cigarettes a day during the 3 months before pregnancy, those with a greater frequency of use had a lower likelihood of quitting e-cigarette use only, including 6–10 cigarettes a day (29.4% vs. 12.5%; 0.82 [0.78–0.85]) and ≥11 cigarettes a day (38.2% vs. 12.5%; 0.73 [0.70–0.76]). Compared to mothers who did not use hookah around the time of pregnancy, mothers who used hookah had a greater likelihood to quit e-cigarette use only (20.7% vs. 29.0%; 1.38 [1.25–1.52]).

Discussion

Changes in e-cigarette use

Similar to previous studies, including in PRAMS,^7–11 we found the prevalence of cigarette use was high before pregnancy, decreased sharply during late pregnancy, and raised moderately after pregnancy. A decreased return-to-use pattern was visible among mothers who exclusively used e-cigarettes and those who used both cigarettes and e-cigarettes, such that use decreased in both groups during late pregnancy from their respective prepregnancy rates, and then increased after pregnancy to rates higher than during late pregnancy, but still below their prepregnancy level. Pregnant women may be encouraged to quit using cigarettes and/or e-cigarettes to avoid adverse birth outcomes.^16,33–35 To our knowledge, there is no existing research on reasons for postpartum return-to-use of e-cigarettes only or dual use of cigarettes and e-cigarettes. However, common reasons (e.g., low motivation or confidence to remain abstinent, partner smoking) for postpartum return-to-use of cigarettes may also be applicable to these findings.³⁶

Most mothers reported no e-cigarette use during the last 3 months of pregnancy, regardless of their frequency of e-cigarette use before pregnancy. The next most common outcome, although much less frequent overall, was staying in the same category of e-cigarette use frequency, followed by a decrease or an increase in use. The prevalence of quitting among mothers who exclusively used e-cigarettes was higher in our study compared to the prevalence of cigarette smoking cessation in a previous PRAMS study.¹⁰

Predictors of quitting exclusive e-cigarette use

As we expected, most mothers who exclusively used e-cigarettes quit use during late pregnancy, particularly those using once a day or less before pregnancy. We found that the youngest women were most likely to quit exclusive e-cigarette use compared to older age groups. According to previous literature, this may be possibly explained by younger women having lower levels of nicotine dependence or more commonly having smoke-free homes.³⁷ Mothers with a college degree or more were more likely to quit e-cigarette use. Possible explanations for this finding may include better health literacy³⁸ and less exposure to tobacco advertisements.³⁹

We found married women were also more likely to quit e-cigarette use. According to previous literature, this may be due to increased support from their partners.^40–42 In our study, mothers who were underweight, overweight, or obese were less likely to quit e-cigarette use. Underweight mothers may be less likely to quit e-cigarettes due to weight concerns or symptoms of serious eating disorders.⁴³ For those with an overweight BMI, having an overweight status has been previously identified as a predictor of using e-cigarettes for weight management.⁴⁴

Changes in dual use

Similar to our study, previous work in PRAMS has shown that some expectant mothers who used both cigarettes and e-cigarettes before their pregnancy quit at least one product during their pregnancy, and a greater proportion quit both.¹⁹ Compared to those who quit e-cigarette use only, we found a relatively small proportion of women quit cigarette use only. Although reasons remain unclear, it may be due to apprehension about e-cigarettes among pregnant women. One previous study found that the novelty of e-cigarettes made some pregnant women believe the health effects were poorly understood.⁴⁵

In addition, device malfunction and ignition and lack of familiarity with the ingredients sparked numerous safety concerns.⁴⁵ Furthermore, in 2019, safety concerns over e-cigarette use in the United States arose after reports of e-cigarette- or vaping product-associated lung injury (EVALI) spread, which caused a decrease in e-cigarette use.⁴⁶ The CDC then released guidance discouraging the use of e-cigarette and vaping products containing vitamin E acetate, especially among pregnant women.⁴⁷ While EVALI is likely due to the use of vaping products containing vitamin E acetate,⁴⁷ and not specifically due to nicotine in vaping products, some people who vape may have chosen to avoid vaping products altogether as a result of these concerns, which possibly make them more likely to quit e-cigarette use.

Predictors of quitting dual use

As with mothers who exclusively used e-cigarettes, the youngest mothers who reported dual use in our study were more likely to quit both than their older counterparts. This is consistent with a previous study, where younger age was associated with less nicotine dependence and a higher prevalence of smoke-free homes.³⁷ The oldest mothers who reported dual use were most likely to quit e-cigarette use only. Although there is no clear reason in the literature to explain this finding, we suspect older pregnant women with longer lifetime cigarette use compared to e-cigarettes may possibly have less difficulty to quit e-cigarettes.

Similar to mothers who exclusively used e-cigarettes, married mothers who reported dual use in our study were more likely to quit both, which may be due to increased support and accountability within the home, according to previous literature on cigarette use.^40–42 Our study also found associations between higher education and increased likelihood of quitting cigarette, e-cigarette, and dual use, aligning with previous studies.^48,49

College graduates who reported dual use were previously reported to have a higher likelihood of quitting both cigarettes and e-cigarettes.⁴⁹ For income, most studies had similar findings to ours. In one study, those who reported dual use and had higher household income were more likely to quit cigarette use only, while continuing e-cigarette use.⁴⁹ According to previous literature, those with lower income were less likely to report e-cigarette use as a reason to reduce their cigarette usage.^49,50 Cigarette smokers in another study with household income at ≥300% of the federal poverty line had greater odds of 30+ day abstinence.⁵¹

According to our findings, mothers who had nongovernmental health insurance or had a higher number of prenatal care visits were associated with a greater likelihood of all quitting behaviors. More prenatal care visits might be an indicator of health-conscious mothers seeking further examinations, including those who need urgent care for a more difficult pregnancy⁵² or mothers who generally have increased health care use.⁵³

Compared to mothers with a normal prepregnancy BMI, obese mothers were positively associated with all quitting behaviors; underweight was inversely associated and overweight was bidirectional. Similar to mothers who exclusively used e-cigarettes, underweight mothers using e-cigarettes and/or cigarettes may be more hesitant to quit due to serious eating disorder symptomatology or weight concerns.⁴³ Overweight mothers may be less motivated to quit e-cigarette use only and more likely to quit cigarette use only as a result of pursuing a weight control option that involves less nicotine.⁴⁴ In contrast, obese mothers may consider no additional benefit from quitting one over the other.

Mothers with one or more previous live births had a lower likelihood of all quitting behaviors. This suggests that those with previous, healthy live births may have a lower risk perception of e-cigarettes and/or cigarettes compared to first-time mothers, consistent with previous literature.⁵⁴ In contrast, we suspect first-time mothers may be more health conscious and/or anxious about ensuring a positive pregnancy, increasing their motivation to quit.

Hispanic mothers were more likely to quit cigarettes only and less likely to quit e-cigarettes only than non-Hispanic mothers. Hispanics have been associated with higher rates of successful cigarette smoking cessation, including during the early phases of pregnancy,⁴² without help from a physician, instead relying more on self-motivation,⁵⁵ and possible self-help methods that are culturally relevant.⁵⁶ Furthermore, non-US-born Hispanic women have higher rates of quitting cigarettes than US-born Hispanic women,⁵⁷ with many reporting smoking within homes as culturally unacceptable.⁵⁸

American Indian or Alaskan Native mothers were least likely to quit both cigarettes and e-cigarettes compared to White mothers. This may be possibly a result of widespread cigarette use in their environments, and potentially weak tobacco control on tribal lands.⁵⁹ Although Asian mothers were less likely than White mothers to quit e-cigarette use only, they were more likely to quit cigarette use only, similar to previous cigarette smoking cessation rates.⁷ According to the literature, Asian/Pacific Islander smokers during pregnancy had increased odds of hypertension/preeclampsia/eclampsia compared to White smokers,⁶⁰ and this may have led to possible rigorous counseling on cigarette smoking cessation. Furthermore, religious practices among Asian communities may impact their use of tobacco as well.⁶¹

Strengths and limitations

Our analysis had several strengths. First, our study was one of the first to examine quitting dual use around the time of pregnancy. Second, we used the most recent Phase 8 data from PRAMS, one of the largest national surveys in the United States. Although PRAMS was only conducted within the United States, our findings were potentially generalizable to other developed countries with similar prevalence of tobacco use during pregnancy. Finally, we used the sample weights generated by the CDC to control for nonresponse bias and nonrandom sampling.

However, our analysis also had some limitations. First, the self-reported status of e-cigarette and cigarette use was subject to recall bias, causing potential misclassification. Second, the PRAMS did not collect postpartum data on e-cigarette use nationally. Our statistical power was limited by the relatively low prevalence of e-cigarette use.

Moreover, our analysis was limited by the lack of information on the dose of e-cigarette use and e-cigarette product information, including flavors, nicotine concentration, and devices. Cigarette use and e-cigarette use were asked about using different questions in the PRAMS, and we binned the variables differently for analysis, meaning direct comparison between cigarette and e-cigarette use frequency is not possible in this study. Furthermore, changes in e-cigarette use may be underestimated here due to the broad categories of use captured in the PRAMS. More specific data collection regarding different levels of use would allow more detailed trend analysis. We also could not examine the predictors for reduction or increase in the frequency of e-cigarette use due to the low proportions of these changes.

Since PRAMS only collected data regarding the 3 months before pregnancy and the last 3 months of pregnancy, we were also limited in our ability to fully assess changes during the first and second trimesters. Some trends occurring during early pregnancy could be missing from our analysis due to this lack of data. Future studies should investigate changes throughout the entirety of pregnancy. Finally, since the time of data collection preceded the onset of the COVID-19 pandemic, which broadly altered attitudes toward health behaviors, it is unclear if these data will be generalizable to current expectant mothers. Reducing the risk from COVID-19 infection was a commonly reported reason to quit among people who reported dual use.⁶² However, assessments before, during, and after the pandemic are needed.

Conclusions

Within 2016–2019 PRAMS data, we found that among mothers who reported dual use, most sociodemographic and pregnancy characteristics predicted quitting use of e-cigarettes and/or cigarettes during pregnancy. However, postpartum return to use of e-cigarettes remains a public health challenge, similar to cigarettes. Our findings may inform programs assisting pregnant women in quitting dual use. However, given that our analyses were exploratory, future research is needed to replicate our results in larger cohorts with more precise quantification of e-cigarette use. In addition, obtaining detailed information on dose, patterns, and products of dual use, and examining health outcomes are warranted.

Ethics Approval/Exemption Statement

This secondary data analysis used the de-identified PRAMS data provided by the CDC. It was approved by the University at Buffalo Institutional Review Board as nonhuman-subjects research. Informed consent from study participants was not required.

Authors' Contributions

X.W.: Conceptualization, guided the plan of data analysis, and co-drafted/revised the article. N.N.: Conducted the statistical analysis, contributed to the literature review, co-generated Tables and Figures, drafted the Abstract and Results sections, and co-drafted the Introduction, Materials and Methods, and Discussion sections. A.M.: Contributed to the literature review, co-drafted the Discussion and Conclusions section, and co-generated Figures. S.R.: Contributed to the literature review, co-drafted the Introduction section, and co-drafted the Discussion section. O.O.: Contributed to the statistical analysis, co-generated Tables and Figures, and co-drafted the Materials and Methods section. E.A.: Contributed to the literature review, interpretation of results, and co-drafted the Discussion section. P.C.: Contributed to the literature review, interpretation of results, and co-drafted the Introduction section.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported, in part, through research support from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), National Institutes of Health (NIH) grant R21HD091515; National Institute on Drug Abuse (NIDA) and the Food and Drug Administration (FDA) Center for Tobacco Products (CTP), NIH grant R21DA053638 (both awarded to Xiaozhong Wen). The information, content, and/or conclusions are those of the author and should not be construed as the official position or policy of, nor should any endorsement be inferred by NIH, FDA, HHS, or the US Government. The sponsors had no role in writing the article or the decision to submit it for publication.

Supplementary Material

Supplementary Table S1

Supplementary Figure S1