Perinatal Smoking and E-cigarette Use and Their Relationship with Breastfeeding: PRAMS 2015

Abstract

Introduction:

Electronic cigarette (e-cigarette) use poses concerns among women of child-bearing age. We examined prevalence of breastfeeding among the U.S. women and characteristics associated with exclusive use of cigarettes, e-cigarettes, or both products.

Materials and Methods:

Our study is based on pooled cross-sectional data from 2015–2020 Pregnancy Risk Assessment Monitoring System. Two outcomes were breastfeeding initiation and breastfeeding duration for over 6 months, as recommended by the American Academy of Pediatrics (AAP). Binary logistic regressions were used to examine associations between each outcome and type of tobacco products unadjusted and adjusted for potential confounders, with post hoc estimation of average adjusted predictions, marginal effects, and contrasts of margins in Stata.

Results:

Adjusted prevalence of breastfeeding initiation was significantly higher in women who used e-cigarettes (86.15%) than conventional cigarettes (72.16%) or both products (79.54%). Similarly, a significantly higher percentage of women who used e-cigarettes continued breastfeeding after 6 months (49.20%) than women who smoked conventional cigarettes (31.30%) or both products (29.83%). Among women who neither smoked nor used e-cigarettes, 85.29% initiated breastfeeding and 57.20% continued breastfeeding as recommended by the AAP.

Conclusion:

Likelihood of breastfeeding initiation and continuation in women using e-cigarettes was comparable to those who neither smoked nor used e-cigarettes. Future research needs to elucidate differences in breastfeeding by sociodemographic and health-related characteristics of women who smoke or use both products compared to those who use e-cigarettes. Understanding women's motivation behind use of a particular tobacco product is also important, so mothers who smoke e-cigarettes are not incorrectly perceived as safer users and excluded from tobacco cessation interventions.

Introduction

Despite breastfeeding being the optimal form of feeding for the healthy development of infants and the well-documented benefits for both babies and their mothers, breastfeeding rates remain variable. In 2012, nearly 77% of infants were ever breastfed and 16% of infants were exclusively breastfed for 6 months. Breastfeeding rates reached the highest peak in 15 years in 2020, when 84% of infants were ever breastfed and nearly 26% of infants were exclusively breastfed for 6 months. Following this peak, there was a slight decline in both breastfeeding initiation and 6-month duration of exclusive breastfeeding in 2022, 83% and 25%, respectively.^1,2 Although there has been an overall steady increase in breastfeeding initiation rates over time, most mothers do not continue exclusive breastfeeding for 6 months, which is the American Academy of Pediatrics (AAP) recommended timeframe.^1,3

Maternal smoking is an established behavioral practice associated with reduced breastfeeding initiation and early breastfeeding cessation.^4–7 Use of electronic cigarettes (e-cigarettes) or vaping is a growing phenomenon among women of childbearing age, and it carries many of the same risks as conventional smoking, such as impaired brain and lung development of exposed infants.^8–10 Research suggests women's knowledge and perceptions regarding concurrent breastfeeding and smoking could explain the difference in breastfeeding outcomes between women who smoke and those who do not. A cross-sectional study conducted in collaboration between a pediatric clinic and large academic medical center in Pittsburgh found that 80% of mothers in their sample had believed that women should not use any tobacco products and breastfeed. Less than 20% of women were aware of the recommendation to continue breastfeeding regardless of smoking status.^11,12

When e-cigarettes were introduced in 2003, they were marketed as a safer alternative to commercial tobacco cigarettes and a mechanism to reduce conventional cigarette addiction. These initial claims have resulted in the growing use of the smokeless devices, even among pregnant and nursing mothers.¹³ E-cigarettes have not been used for long enough time to fully understand the effects of their use by pregnant and breastfeeding mothers on the health of infants. Research has, however, documented the harmful health risks associated with nicotine and infants. E-cigarette aerosols may contain nicotine as well as other harmful chemicals, including flavorings, toxicants, and solvents that can alter the healthy development of children.^14,15

Research is not clear as to why maternal smoking is a strong predictor of low intention of breastfeeding initiation and early breastfeeding cessation. The intent to breastfeed, among mothers who smoke and/or use e-cigarettes, may be influenced by the perception of the smoking product and the impact the behavior may have on the infant. Individuals often behave according to their perceptions.¹⁶ Prior studies suggest that many mothers perceive e-cigarettes to be safer than combustible tobacco cigarettes. This perception may contribute to a difference in breastfeeding outcomes among those who smoke tobacco products and use e-cigarettes.¹⁷

Based on the groundless presumption of e-cigarettes being less harmful for mothers and their babies, it is reasonable to assume higher prevalence of breastfeeding among e-cigarette users than conventional smokers. Hence, the purpose of our study was to examine difference in breastfeeding initiation and duration by type of tobacco products, specifically, cigarettes alone, e-cigarettes alone, or dual use of electronic and combustible cigarettes among mothers.

To date, no study has compared the difference in breastfeeding outcomes between mothers who smoke versus those who used e-cigarettes. Understanding such differences and associated factors will help identify next steps for addressing misinformation on e-cigarettes as a safer alternative to cigarettes for women and their babies and groups of women in need for interventions.

The AAP encourages mothers who use commercial tobacco products and e-cigarettes to breastfeed. To minimize the transmission of nicotine and other by-products to the infant, the AAP indicates that it is preferable that the mother smoke or vape immediately after breastfeeding.¹⁸ Breastfeeding provides protection to newborns against potential chemical exposures. Many women and providers are unaware of the recommendations and have negative attitudes toward the breastfeeding by mother who use tobacco products. Following the current AAP recommendations regarding a mother's smoking and breastfeeding could increase overall breastfeeding rates.

Concurrent breastfeeding and smoking are viewed as superior for infant health than formula feeding and maternal smoking because infants are likely to still be exposed to the chemicals through secondhand smoke, which may place the child at greater risk for respiratory illness, allergic disorders, and sudden infant death. Mothers who smoke conventional cigarettes and/or e-cigarettes should be encouraged to quit. They should also still be encouraged to breastfeed. The benefits of breastfeeding to both mothers and their babies far outweigh the potential risk of exposure to nicotine and other harmful chemicals found in breast milk after smoking and e-cigarette use.¹⁸ Further, mothers who use e-cigarettes and incorrectly perceive such products as safe should be educated on the associated health risks and included in smoking cessation interventions.

Materials and Methods

Data source

This was a cross-sectional study based on pooled data from 2015–2020 National Pregnancy Risk Assessment Monitoring System (PRAMS) survey. PRAMS is an ongoing, population-based surveillance system introduced by the Centers for Disease Control and Prevention and state health departments in 1987 to monitor maternal health and behaviors before, during, and after pregnancy.¹⁹ The purpose of PRAMS is to decrease morbidity and mortality of mothers and infants and improve their health through the reduction of adverse events relative to pregnancy and postpartum.²⁰ The survey is conducted during the postpartum period in 37 states and New York City using a mixed mode data collection and stratified sampling technique with an oversampling on characteristics about high-risk women.

States participating in PRAMS use a standardized core questionnaire, and they can also add questions related to their own interests. The scope of PRAMS survey encompasses prenatal care, attitudes, and feelings about previous pregnancy, health insurance coverage, cigarette smoking, drinking, physical abuse, maternal stress, economic status, and infant health status.²⁰ Each participating site samples approximately between 1,000 and 3,000 women each year. Women are contacted 2–6 months after giving birth to a live baby. The survey is conducted both in English and in Spanish. The final version of PRAMS data is a combination of survey data and data from infants' birth certificates.

PRAMS has a minimum overall weighted response rate threshold policy for the data release each year. The threshold was 55% for 2015 and 2017 and 50% for 2018–2021. Between 2015 and 2020, weighted response rates ranged from 33.7% to 81.3%. We started with PRAMS 2015 as this was the first survey year when data on e-cigarettes use had started to be collected.¹⁹ Our final analytic sample consisted of 247,537 observations collected between 2015 and 2020 in 49 jurisdictions.

Measures

The two outcomes of interest were breastfeeding initiation and breastfeeding duration. Breastfeeding initiation was operationalized using responses to the question “Did you ever breastfeed or pump milk to feed your new baby, even for a short period of time?” Those who answered “yes” were considered to have initiated breastfeeding during the postpartum period. The breastfeeding duration was defined using responses on how many weeks or months a woman had breastfed or fed pumped milk to her baby. We chose 6 months of exclusive breastfeeding duration as our cutoff point for the binary variable on breastfeeding duration consistent with the WHO recommendation on the exclusive breastfeeding duration and the AAP-updated guidelines extending duration of breastfeeding to 6 months from 4 months.^3,21

Our main independent variable, use of tobacco products was operationalized using women's responses on whether they smoked and whether they used e-cigarettes in last 2 years. Answers from both questions were used to differentiate between women who used none of the products, cigarettes exclusively, e-cigarettes exclusively, or both products. Potential confounding variables identified by prior research, included race, ethnicity, household income, the body mass index, alcohol drinking, prenatal visits, education levels, type of delivery (vaginal or not), mother's receipt of food assistance, mother's attempt to become pregnant, mother having a conversation with health care workers about breastfeeding, previous live births, previous deliveries, and marital status.^7,18,22,23

Statistical analysis

All analysis were weighted to account for the complex survey design and unit non-response. The analyses were conducted using StataBE 17.0. Weighted percentages and corresponding 95% confidence intervals (CIs) were reported for the categorical variables overall and by breastfeeding initiation and duration. Unadjusted associations between each of the outcomes and independent variables were examined using univariate logistic regressions. Binary logistic regressions were used to examine associations between breastfeeding initiation and duration and type of tobacco products, adjusting for the remaining independent variables. Given high prevalence of the outcome (>20%) and importance of making results more tangible, all logistic regression analyses were followed by computing average adjusted predictions and marginal effects and estimating contrasts of margins.^24,25

Given high collinearity between income categories and participation in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) and higher percentage of missing observations for the former variable, we used the latter in the multivariable logistic regression analyses. Also, given high percentage of observations missing data on conversation with a health care worker about breastfeeding and trying to get pregnant (Tables 1 and 2), we excluded these variables from the fully adjusted models (Table 3). However, in sensitivity analyses with these variables included, reported associations and statistical effect on outcome from use of different tobacco products were qualitatively similar to reported ones. All tests were two-tailed. The probability of rejecting the null hypothesis when it is true was set at alpha = 0.05.

Table 1.

Prevalence of Breastfeeding Initiation Among Women: 2015–2020 Pregnancy Risk Assessment Monitoring System

Maternal characteristics	Unweighted, n	Did not initiate breastfeeding % (95% CI) 16.43 (16.20–16.66)	Initiated breastfeeding % (95% CI) 83.57 (83.34–83.8)	P-value
Smoking status^a
Neither	172,217	14.32 (14.06–14.57)	85.68 (85.43–85.94)
Combustible cigarettes	11,819	39.58 (38.06–41.14)	60.42 (58.86–61.94)	<0.001
E-cigarettes	9,341	16.29 (15.14–17.51)	83.71 (82.49–84.86)	0.001
Both	4,185	30.57 (28.32–32.91)	69.43 (67.09–71.68)	<0.001
Age
<19 years	11,527	27.82 (26.4–29.29)	72.18 (70.71–73.60)
20–34 years	184,428	16.55 (16.29–16.82)	83.45 (83.18–83.71)	<0.001
>35 years	44,389	13.03 (12.56–13.52)	83.57 (83.34–83.80)	<0.001
Race
White	143,778	14.77 (14.49–15.05)	85.23 (84.95–85.51)
Black	45,764	27.45 (26.79–28.12)	72.55 (71.88–73.21)	<0.001
Other^b	45,668	12.60 (12.09–13.14)	87.40 (86.86–87.91)	<0.001
Ethnicity
Non-Hispanic	188,977	17.48 (17.22–17.74)	82.52 (82.26–82.78)
Hispanic	45,096	12.26 (11.71–12.83)	87.74 (87.17–88.29)	<0.001
Education
<12 years	30,010	26.10 (25.25–26.96)	73.90 (73.04–74.75)
12 years	58,344	24.57 (23.99–25.15)	75.43 (74.85–76.01)	0.004
13–15 years	68,360	16.40 (15.96–16.85)	83.60 (83.15–84.04)	<0.001
16 or more years	81,595	7.40 (7.149–7.666)	92.60 (92.33–92.85)	<0.001
Marital status
No	97,829	25.08 (24.62–25.54)	74.92 (74.46–75.38)
Yes	142,343	10.97 (10.72–11.21)	89.03 (88.79–89.28)	<0.001
Household income
Less than 25,000	64,937	25.44 (24.89–26.0)	74.56 (74.00–75.11)
25,000–50,000	37,488	15.35 (14.78–15.95)	84.65 (84.05–85.22)	<0.001
51,000–75,000	25,443	12.14 (11.53–12.77)	87.86 (87.23–88.47)	<0.001
>75,000 or more	57,210	7.71 (7.39–8.04)	92.29 (91.96–92.6)	<0.001
Receipt of WIC food during pregnancy
No	144,337	11.92 (11.67–12.18)	88.08 (87.82–88.33)
Yes	92,582	24.20 (23.73–24.66)	75.80 (75.34–76.27)	<0.001
Body mass index
Underweight	8,362	19.91 (18.48–21.41)	80.09 (78.59–81.52)
Normal	101,669	13.84 (13.52–14.17)	86.16 (85.83–86.48)	<0.001
Overweight	58,226	16.11 (15.62–16.6)	83.89 (83.40–84.38)	<0.001
Obese	60,987	20.74 (20.22–21.26)	79.26 (78.74–79.78)	0.296
Vaginal delivery
No	85,067	18.30 (17.88–18.73)	81.70 (81.27–82.12)
Yes	155,163	15.51 (15.23–15.79)	84.49 (84.21–84.77)	<0.001
Previous live birth
No	93,951	13.44 (13.10–13.79)	86.56 (86.21–86.9)
Yes	145,985	18.33 (18.01–18.65)	81.67 (81.35–81.99)	<0.001
Plural birth
No	227,014	16.33 (16.09–16.56)	83.67 ((83.44–83.91)
Yes	8,202	23.24 (21.63–24.93)	76.76 (75.07–78.37)	<0.001
Drink the last 3 months of pregnancy
No	33,819	18.61 (17.93–19.31)	81.39 (80.69–82.07)
Yes	2,661	11.45 (9.74–13.43)	88.55 (86.57–90.26)	<0.001
Prenatal care within the first trimester
No	32,402	23.40 (22.62–24.2)	76.60 (75.80–77.38)
Yes	202,037	15.18 (14.94–15.42)	84.82 (84.58–85.06)	<0.001
Health care worker talked about breastfeeding
No	11,382	14.56 (13.58–15.59)	85.44 (84.41–86.42)
Yes	61,399	17.69 (17.24–18.15)	82.31 (81.0–82.76)	<0.001
Trying to get pregnant
No	89,045	21.37 (20.95–21.81)	78.63 (78.19–79.05)
Yes	105,034	12.43 (12.13–12.74)	87.57 (87.26–87.87)	<0.001

Crude prevalence, reported as %, corresponding 95% CIs, and p-values are based on calculated average adjusted predictions, following univariate logistic regressions.

Grouped variable created in PRAMS was used to assess smoking. The variable combined responses to three questions about maternal smoking before, during, or after pregnancy.

Other includes Asian, Chinese, Japanese, Filipino, Hawaiian, Alaskan Native, American Indian, mixed race, and other nonwhite.

CI, confidence interval; PRAMS, Pregnancy Risk Assessment Monitoring System; WIC, The Special Supplemental Nutrition Program for Women, Infants, and Children.

Table 2.

Prevalence of Breastfeeding Continuation for over 6 Months Among Mothers: 2015–2020 Pregnancy Risk Assessment Monitoring System

Maternal characteristics	Unweighted, n	Breastfed <6 months % (95% CI) 46.09 (45.77–46.42)	Breastfed 6+ months % (95% CI) 53.91% (53.5–54.23)	P-value
Smoking status^a
Neither	168,980	42.61 (42.23–42.98)	57.39 (57.02–57.77)
Combustible cigarettes	11,187	79.51 (78.18–80.78)	20.49 (19.22–21.82)	<0.001
E-cigarettes	9,079	55.67 (54.02–57.3)	44.33 (42.70–45.98)	<0.001
Both	3,920	76.87 (74.65–78.96)	23.13 (21.04–25.35)	<0.001
Age
<19 years	11,175	75.59 (74.13–76.99)	24.41 (23.01–25.87)
20–34 years	182,085	46.40 (46.03–46.77)	53.60 (53.23–53.97)	<0.001
>35 years	43,951	37.51 (36.79–38.24)	62.40 (61.76–63.21)	<0.001
Race
White	141,061	43.07 (42.68–43.47)	56.93 (56.53–57.32)
Black	43,914	63.35 (62.59–64.11)	36.65 (35.89–37.41)	<0.001
Other	43,969	42.80 (41.94–43.65)	57.20 (56.35–58.06)	0.564
Ethnicity^b
Non-Hispanic	186,634	45.16 (44.81–45.51)	54.84 (54.49–55.19)
Hispanic	44,444	50.52 (49.67–51.38)	49.48 (48.62–50.33)	<0.001
Education
<12 years	28,835	61.88 (60.86–62.88)	38.12 (37.12–39.14)
12 years	56,783	63.27 (62.59–63.94)	36.73 (36.06–37.41)	0.025
13–15 years	67,718	50.91 (50.29–51.52)	49.09 (48.48–49.71)	<0.001
16 or more years	81,941	25.74 (25.28–26.2)	74.26 (73.80–74.72)	<0.001
Marital status
No	94,674	65.72 (65.2–66.24)	34.28 (33.76–34.81)
Yes	141,936	34.01 (33.63–34.4)	65.99 (65.5–66.37)	<0.001
Household income
Less than 25,000	62,778	64.42 (63.78–65.06)	35.58 (34.94–36.22)
25,000–50,000	36,920	48.34 (47.51–49.18)	51.66 (50.82–52.49)	<0.001
50,000–75,000	25,194	37.48 (36.54–38.43)	62.52 (61.57–63.46)	<0.001
>75,000	56,827	26.97 (26.42–27.52)	73.03 (72.48–73.58)	<0.001
Receipt of WIC food during pregnancy
No	141,271	36.84 (36.45–37.24)	63.16 (62.76–63.55)
Yes	89,298	62.57 (62.01–63.11)	37.43 (36.89–37.99)	<0.001
Body mass index
Underweight	8,220	50.42 (48.58–52.25)	49.58 (47.75–51.42)
Normal	100,726	38.69 (38.21–39.17)	61.31 (60.83–61.79)	<0.001
Overweight	57,504	46.05 (45.39–46.72)	53.95 (53.28–54.61)	<0.001
Obese	60,172	58.35 (57.69–59.00)	41.65 (41.0–42.31)	<0.001
Vaginal delivery
No	83,820	51.12 (50.55–51.68)	48.88 (48.32–49.45)
Yes	153,234	43.64 (43.24–44.04)	56.36 (55.96–56.76)	<0.001
Previous live birth
No	92,652	46.70 (46.18–47.23)	53.30 (52.77–53.82)
Yes	143,636	45.75 (45.34–46.17)	54.25 (53.83–54.66)	0.006
Plural birth
No	224,093	45.93 (45.61–46.26)	54.07 (53.74–54.39)
Yes	8,121	57.33 (55.41–59.24)	42.67 (40.76–44.59)	<0.001
Drink the last 3 months of pregnancy
No	36,463	49.01 (48.12–49.9)	50.99 (50.10–51.88)
Yes	2,885	36.16 (33.37–39.05)	63.84 (60.95–66.63)	<0.001
Prenatal care within the first trimester
No	31,294	55.63 (54.65–56.6)	44.37 (43.40–45.35)
Yes	200,780	44.47 (44.12–44.81)	55.53 (55.19–55.88)	<0.001
Health care worker talked about breastfeeding
No	11,644	38.84 (37.49–40.21)	61.16 (59.79–62.51)
Yes	62,960	50.66 (50.05–51.28)	49.34 (48.72–49.95)	<0.001
Trying to get pregnant
No	87,641	56.73 (56.19–57.27)	43.27 (42.73–43.81)
Yes	104,700	38.02 (37.56–38.48)	61.98 (61.52–62.44)	<0.001

Crude prevalence, reported as %, corresponding 95% CIs, and p-values are based on calculated average adjusted predictions, following univariate logistic regressions.

Grouped variable created in PRAMS was used to assess smoking. The variable combined responses to three questions about maternal smoking before, during, or after pregnancy.

Other includes Asian, Chinese, Japanese, Filipino, Hawaiian, Alaskan Native, American Indian, mixed race, and other nonwhite.

CI, confidence interval; PRAMS, Pregnancy Risk Assessment Monitoring System; WIC, The Special Supplemental Nutrition Program for Women, Infants, and Children.

Table 3.

Adjusted Association Between Breastfeeding Initiation and Continuation for over 6 Months Among Mothers: 2015–2020 Pregnancy Risk Assessment Monitoring System

Smoking status	Breastfeeding initiation				Breastfeeding continuation for 6+ months
Smoking status	% (95% CI) Adjusted for sociodemographic characteristics^a	P-value	% (95% CI) Adjusted for all characteristics^b	P-value	% (95% CI) Adjusted for sociodemographic characteristics^a	P-value	% (95% CI) Adjusted for all characteristics^b	P-value
Neither	84.84(84.57–85.11)	Ref.	84.91(84.63–85.19)	Ref.	55.82^***(55.46–56.19)	Ref.	56.00^***(55.63–56.37)	Ref.
Combustible	73.49^***(72.24–74.74)	<0.001	74.67^***(73.42–75.29)	<0.001	33.89^***(32.18–35.60)	<0.001	34.78^***(33.02–36.53)	<0.001
E-cigarettes	86.41^***(85.39–87.43)	0.004	86.26^***(85.22–87.31)	0.015	51.44^***(49.81–53.07)	<0.001	52.25^***(50.63–53.87)	<0.001
Both	79.34^***(77.60–81.08)	<0.001	80.19^***(78.46–81.93)	<0.001	36.67^***(34.03–39.31)	<0.001	38.00^***(35.32–40.67)	<0.001

Reported % are average adjusted predictions, calculated following the binary logistic regressions, adjusted for age, race, ethnicity, education, marital status, and receipt of WIC food during pregnancy; n's = 189,179 and 184,943.

Reported % are average adjusted predictions, calculated following the binary logistic regressions, adjusted for the above sociodemographic characteristics, and body mass index, vaginal delivery, previous live birth, plural birth, and prenatal care within the first trimester; n's = 176,703 and 173,102.

***

p < 0.001 statistically significant differences between two consecutive categories based on estimating contrasts of margins.

CI, confidence interval; WIC, The Special Supplemental Nutrition Program for Women, Infants, and Children.

The PRAMS data are deidentified and publicly available; hence, its analysis does not constitute human subject research. The investigators did not have to obtain informed consent. The study was exempt from review by the Research Ethics Boards at the investigators' institutions.

Results

Prevalence of breastfeeding initiation was 83.57% (95% CI: 83.34–83.8) among women with live-born infants between 2015 and 2020. Slightly over half of women breastfed their infants for at least 6 months (53.91%; 95% CI: 53.5–54.23). Similar percentages of women reported smoking combustible cigarettes exclusively (4.98%; 95% CI: 4.83–5.13) or e-cigarettes exclusively (4.95%; 95% CI: 4.79–5.11), and 1.88% of women smoked both products (95% CI: 1.79–1.98). Each sociodemographic and health-related characteristics of mothers was significantly associated with breastfeeding initiation and continuation over 6 months, as summarized in Tables 1 and 2.

As shown in Table 1, in crude analyses, 60.42% of women who smoked cigarettes, initiated breastfeeding (95% CI: 58.87–62.53%), compared to 83.71% (95% CI: 82.53–84.89%) of women who used e-cigarettes or 85.68% (95% CI: 85.43–85.94%) of women who neither smoked nor vaped (all p's < 0.001). The percentage of women who had initiated breastfeeding and used both products, was similar to those who smoked cigarettes: 69.43% (95% CI: 67.14–71.76). Likewise, as shown in Table 2, a significantly lower percentage of women who smoked cigarettes (20.49%; 95 CI: 19.19–21.79) and those who used both products (23.13%; 95% CI: 20.97–25.28) breastfed their infant after 6 months compared to women who used e-cigarettes (44.33%; 95 CI: 42.69–45.97) or did not use either product (57.39%; 95% CI: 57.02–57.77).

When comparing adjusted (Table 3) versus crude (Table 1) estimates, among women who smoked cigarettes, prevalence of breastfeeding initiation adjusted for sociodemographic characteristics (73.49%; 95% CI: 72.24–74.74) was higher by 13.07 percentage points than the crude prevalence of 60.42% (95% CI: 58.86–61.94); and prevalence of breastfeeding initiation also adjusted for health-related characteristics (74.67%; 95% CI: 73.42–75.29) was higher by 14.25 percentage points than the crude prevalence reported in Table 1. Second largest differences between adjusted and crude prevalence estimates were among women who reported both smoking cigarettes and using e-cigarettes: higher by 9.91 percentage points controlling for sociodemographic characteristics only and by 10.76 percentage points controlling for additional health-related characteristics (Table 3) versus crude prevalence of 69.43% (95% CI: 67.09–71.68) (Table 1).

Among women who used e-cigarettes, adjusted prevalence of breastfeeding initiation was just under 3 percentage points higher (Table 3) than the crude prevalence of 83.71% (95% CI: 82.49–84.86) reported in Table 1. Among women who neither smoked nor vaped, adjusted prevalence of breastfeeding initiation was lower by less than 1 percentage point (Table 3) than crude prevalence (Table 1).

Among women who smoked cigarettes, prevalence of breastfeeding continuation past 6 months adjusted for sociodemographic characteristics was higher by 13.40 percentage points and adjusted for additional health-related characteristics and was higher by 14.29 percentage points than the crude prevalence of 20.49% (95% CI: 19.22–21.82) reported in Table 2. Other differences between adjusted (Table 3) and crude estimates of prevalence continuation (Table 2) were among women who reported smoking and vaping: higher by 13.54 and 14.87 percentage points controlling for sociodemographic characteristics only and added health-related characteristics, respectively. Among women who used e-cigarettes, adjusted prevalence of breastfeeding continuation was higher by more than 5 percentage points than the crude prevalence of 44.33% (95% CI: 42.7–45.98). Among women who neither smoked nor vaped, adjusted prevalence of breastfeeding continuation was lower by less than 1 percentage point than the crude prevalence of 57.39% (95% CI: 57.02–57.77) reported in Table 2.

Adjusting for sociodemographic and health-related characteristics, among women who used e-cigarettes significantly, higher percentage had initiated breastfeeding (86.26%; 95% CI: 85.22–87.31) than women who neither smoked nor vaped (84.91%; 95% CI: 84.63–85.19) (p = 0.015) and those who smoked cigarettes (74.67%; 95% CI: 73.42–75.29) or used both products (80.19%; 95% CI: 78.46–81.93) (p's < 0.001). Based on fully adjusted analysis of breastfeeding continuation, compared to women who neither smoked nor vaped, those who used e-cigarettes were slightly less likely to breastfeed their infants for at least 6 months (56% versus 52.25%, p < 0.001). Yet, a significantly higher percentage of women who used e-cigarettes (52.25%) continued breastfeeding compared to women who smoked cigarettes (34.78%; 95% CI: 33.02–36.53) or both products (38.00%; 95% CI: 35.32–40.67) (p's < 0.001) (Table 3).

Discussion

To increase breastfeeding and smoking cessation rates, it is important to understand differences in breastfeeding initiation and continuation among women, depending on the type of tobacco product used. Research on such differences has been limited. We have found mothers who smoked exclusively and those who both smoked combustible cigarettes and used e-cigarettes were less likely to initiate breastfeeding and continue it for over 6 months, as recommended by the AAP,³ compared to mothers who neither smoked nor used e-cigarettes. This finding is consistent with prior research documenting prenatal smoking as one of the strongest determinants of established prenatal intent to not breastfeed and reduced breastfeeding duration.^24,25

The association may be partly explained by confounding factors such as sociodemographic characteristics of women (e.g., age, race, education, and marital status) as reflected in the differences of more than 10 percentage points between crude and adjusted prevalence of breastfeeding initiation and continuation after 6 months among women who smoke exclusively or in combination with e-cigarettes. It is also possible that mothers have safety concerns regarding smoking combustible cigarettes more so than other tobacco products.¹¹ Of note, in a cross-sectional study based on data from Jordan's Population and Family Health Surveys 2012 and 2017–2018, mothers who smoked combustible cigarettes were significantly less likely to breastfeed their infants under 25 months compared to women who neither smoked cigarettes nor used a water pipe (AOR 0.51, 95% CI 0.30–0.87).²⁶

We have also found that among women who used e-cigarettes, adjusted prevalence of breastfeeding initiation was 1.35 percentage points higher compared to women who neither smoked nor vaped, 11.59 percentage points higher compared to women who smoked cigarettes, and 6.07 percentage points higher compared to women who smoked and vaped. Although prevalence of breastfeeding continuation was lower by 3.75 percentage points among women who used e-cigarettes compared to those who neither smoked nor vaped, it was significantly higher compared to women who smoked cigarettes (by 17.47 percentage points) and those who smoked and vaped (by 14.25 percentage points). These findings coupled with smaller confounding from sociodemographic and health-related characteristics of the association between breastfeeding behaviors and e-cigarettes use than breastfeeding behaviors and smoking could be attributed to lower perceived health risks of e-cigarettes than combustible cigarettes.

Further, a cross-sectional study based on 2016–2018 PRAMS data reported 0.63 times lower odds of breastfeeding for at least 3 months in women who used e-cigarettes during pregnancy compared with those who did not (95% CI: 0.44–0.89; p = 0.010).⁵ Since mothers who use e-cigarettes are less likely to continue breastfeeding their infants than those who neither smoke nor vape, and are more likely to continue breastfeeding than those who smoke or use both tobacco products, there is value in identifying type of tobacco product used by a woman during perinatal period and tailor breastfeeding recommendation accordingly while encouraging smoking cessation.

E-cigarettes have been marketed as an alternative to help pregnant women quit or reduce the use of tobacco. The marketing and misinformation regarding e-cigarettes have led to their increased use by women of childbearing age. Surveys administered to pregnant women to identify their perceptions on e-cigarettes found that women had perceived e-cigarettes to be less harmful than cigarettes. An increasing number of women of child-bearing age are using e-cigarettes due to the lack of perceived health risks attributable to their use.^5,27–29

E-cigarettes contain nicotine and when inhaled, nicotine enters a mother's blood through her lungs, and then passes into breast milk. Nicotine in the breast milk can affect infant sleep patterns resulting in an increased risk for blood sugar and thyroid issues that can lead to childhood obesity. Nicotine also decreases the milk supply by lowering the levels of prolactin, which is a hormone that stimulates breast milk.^14,15 Depending on presence of nicotine, e-cigarettes can be attributed to electronic nicotine delivery systems or electronic non-nicotine delivery systems. However, several e-cigarette products labeled as not containing nicotine have been shown to contain measurable levels of nicotine.³⁰ Both forms heat a liquid containing flavorings and other chemicals (most commonly water, propylene glycol, glycerin) that produces an aerosol in which the user inhales.

Strengths and limitations

The PRAMS data use standardized surveillance methodology, which is linked to birth certificates. Our analysis is based on PRAMS data and is generalizable to population at large, which is one of the main strengths of our study.

Among study limitations are reliance on self-reported smoking and use of e-cigarettes and related information biases, for example, social desirability bias. Since the PRAMS focuses on mother who have delivered, some women might not report tobacco use, as such behavior is not considered healthy during pregnancy and breastfeeding.³¹ Another limitation is the cross-sectional nature of data, which restricts our ability to make causal inferences about breastfeeding initiation and continuation and tobacco product use.³² Additionally, we only analyzed the use of combustible and e-cigarettes and did not examine use of other legally consumed tobacco products like smokeless tobacco, hookah, and shisha due to small number of observations. Future studies would benefit from considering other types of tobacco product use and combinations thereof in terms of breastfeeding practices.

In conclusion, combustible cigarettes and e-cigarettes alike carry harmful health risks, yet, e-cigarettes are seemingly perceived to be safer, resulting in more mothers being comfortable with vaping and breastfeeding. To understand the underlying perceptions of smoking and breastfeeding, it is important to examine the differences in breastfeeding outcomes among the varying groups of women who use different types of tobacco products quantitively and qualitatively. Research does not yet encourage smoking cessation among pregnant women who use e-cigarettes, despite the potential harmful effects of their use.^5,33 Mothers have been advised to replace conventional smoking with vaping; however, neither product is safe.^34–36

Given the relative modernity of e-cigarettes, the full scope of potential harmful effects of their use on breastfeeding mothers and their infants is not yet fully understood. While mothers should be encouraged to breastfeed, despite their smoking status, they should also be informed about potential risks stemming from e-cigarettes and encouraged to quit using tobacco products altogether.

Patient Consent

The study is based on secondary analysis of publicly available deidentified data from the ongoing state-based surveillance system.

Data Availability

The study used publicly available data.

Footnotes

Authors' Contributions

W.N.H. conceptualized the study and drafted and revised the article. N.M. performed data management and analysis, as well as contributed to article writing, preparation for submission, and revisions. C.K. performed data management and analysis and contributed to article writing. Y.N.T. conceptualized the study, guided data analyses and interpretation, and drafted and revised the article.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

References

CDC. Centers for Disease, Control. Breastfeeding Report Card. United States; 2022. Available from: https://www.cdc.gov/breastfeeding/data/reportcard.htm [Last accessed: May 13, 2023].

World Health Organisation. Breastfeeding: Recommendations. 2023. Available from: https://www.cdc.gov/breastfeeding/data/reportcard.htm [Last accessed: May 13, 2023].

Breastfeeding

, Eidelman

, Schanler

, et al. Breastfeeding and the use of human milk. Pediatrics, 2012; 129(3):e827–e841; doi: 10.1542/peds.2011-3552

Ahluwalia

, Morrow

, Hsia

. Why do women stop breastfeeding? Findings from the pregnancy risk assessment and monitoring system. Pediatrics, 2005; 116(6):1408–1412.

McBride

, Haile

. Association between electronic nicotine delivery systems use and breastfeeding duration. Breastfeed Med, 2021; 16(11):886–893.

Tanda

, Chertok

IRA

, Haile

, et al. Factors that modify the association of maternal postpartum smoking and exclusive breastfeeding rates. Breastfeed Med, 2018; 13(9):614–621; doi:10.1089/bfm.2018.0079

Cohen

, Alexander

, Krebs

, et al. Factors associated with breastfeeding initiation and continuation: A meta-analysis. J Pediatrics, 2018; 203:190.e21–196.e21; doi:10.1016/j.jpeds.2018.08.008

England

, Anderson

, Tong

VTK

, et al. Screening practices and attitudes of obstetricians-gynecologists toward new and emerging tobacco products. Am J Obstet Gynecol, 2014; 211(6):695.e1–e695.e7; doi:10.1016/j.ajog.2014.05.041

Health

UDo

, Committee

HSOHC

. US Department of Health and Human Services oral health strategic framework, 2014–2017. Public Health Rep, 2016; 131(2):242–257.

10.

Nguyen

, Tong

, Marynak

, et al. Perceptions of harm to children exposed to secondhand aerosol from electronic vapor products, styles survey, 2015. Prev Chronic Dis, 2017; 14:E41; doi: 10.5888/pcd14.160567

11.

Bogen

, Davies

, Barnhart

, et al. What do mothers think about concurrent breast-feeding and smoking?. Ambul Pediatr, 2008; 8(3):200–204.

12.

Fitzmaurice

, Abate

, Abbasi

, et al. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 29 cancer groups, 1990 to 2017: A systematic analysis for the global burden of disease study. JAMA Oncol, 2019; 5(12):1749–1768; doi: 10.1001/jamaoncol.2019.2996

13.

Sapru

, Vardhan

, Li

, et al. E-cigarettes use in the United States: Reasons for use, perceptions, and effects on health. BMC Public Health, 2020; 20(1):1518; doi: 10.1186/s12889-020-09572-x

14.

Mennella

, Yourshaw

, Morgan

. Breastfeeding and smoking: Short-term effects on infant feeding and sleep. Pediatrics, 2007; 120(3):497–502; doi: 10.1542/peds.2007-0488

15.

Wen

, Shenassa

, Paradis

. Maternal smoking, breastfeeding, and risk of childhood overweight: Findings from a national cohort. Matern Child Health J, 2013; 17:746–755.

16.

Ferguson

, Bargh

. How social perception can automatically influence behavior. Trends Cogn Sci, 2004; 8:33–39; doi: 10.1016/j.tics.2003.11.004

17.

McCubbin

, Wiggins

, Barnett

, et al. Perceptions, characteristics, and behaviors of cigarette and electronic cigarette use among pregnant smokers. Womens Health Issues, 2020; 30(3):221–229.

18.

Meek

, Noble

. Policy statement: Breastfeeding and the use of human milk. Pediatrics, 2022; 150(1):e2022057988; doi: 10.1542/peds.2022-057988

19.

What is PRAMS? Division of Reproductive Health, National Center for Chronic Disease Prevention and Health Promotion. Available from: https://www.cdc.gov/prams/index.htm [Last accessed: May 13, 2023].

20.

Shulman

, D'Angelo

, Harrison

, et al. The pregnancy risk assessment monitoring system (PRAMS): Overview of design and methodology. Am J Public Health, 2018; 108(10):1305–1313.

21.

Black

. American Academy of Pediatrics Calls for More Support for Breastfeeding Mothers Within Updated Policy Recommendations. 2022. Available from: https://www.aap.org/en/news-room/news-releases/aap/2022/american-academy-of-pediatrics-calls-for-more-support-for-breastfeeding-mothers-within-updated-policy-recommendations/ [Last accessed: May 13, 2023].

22.

Carswell

, Ward

, Vander Weg

, et al. Prospective associations of breastfeeding and smoking cessation among low-income pregnant women. Matern Child Nutr, 2018; 14(4):e12622; doi: 10.1111/mcn.12622

23.

Liu

, Rosenberg

, Sandoval

. Breastfeeding duration and perinatal cigarette smoking in a population-based cohort. Am J Public Health, 2006; 96(2):309–314.

24.

Bailey

, Wright

. Breastfeeding initiation in a rural sample: Predictive factors and the role of smoking. J Hum Lact, 2011; 27(1):33–40; doi: 10.1177/0890334410386955

25.

Horta

, Kramer

, Platt

. Maternal smoking and the risk of early weaning: A meta-analysis. Am J Public Health, 2001; 91(2):304–307; doi: 10.2105/ajph.91.2.304

26.

Can Özalp

, Yalçın

. Is maternal cigarette or water pipe use associated with stopping breastfeeding? Evidence from the Jordan population and family health surveys 2012 and 2017–2018. Int Breastfeed J, 2021; 16(1):1–12.

27.

Wagner

, Camerota

, Propper

. Prevalence and perceptions of electronic cigarette use during pregnancy. Matern Child Health J, 2017; 21(8):1655–1661; doi: 10.1007/s10995-016-2257-9

28.

Mark

, Farquhar

, Chisolm

, et al. Knowledge, attitudes, and practice of electronic cigarette use among pregnant women. J Addict Med, 2015; 9(4):266–272.

29.

Breland

, McCubbin

, Ashford

. Electronic nicotine delivery systems and pregnancy: Recent research on perceptions, cessation, and toxicant delivery. Birth Defects Res, 2019; 111(17):1284–1293; doi: 10.1002/bdr2.1561

30.

WHO. WHO Study Group on Tobacco Product Regulation: Report on the Scientific Basis of Tobacco Product Regulation: Eighth Report of a WHO Study Group. World Health Organization: Geneva; 2021.

31.

Latkin

, Edwards

, Davey-Rothwell

, et al. The relationship between social desirability bias and self-reports of health, substance use, and social network factors among urban substance users in Baltimore, Maryland. Addict Behav, 2017; 73:133–136; doi: 10.1016/j.addbeh.2017.05.005

32.

Lauria

, Lamberti

, Grandolfo

. Smoking behaviour before, during, and after pregnancy: The effect of breastfeeding. ScientificWorldJournal, 2012; 2012:154910.

33.

Haile

, Chertok

IRA

, Haider

. Association between a history of hookah use and breastfeeding duration. Breastfeed Med, 2022; 17(8):678–686.

34.

Johnston

, Campbell

, Coleman

, et al. Safety of electronic cigarette use during breastfeeding: Qualitative study using online forum discussions. J Med Internet Res, 2019; 21(8):e11506; doi: 10.2196/11506

35.

Reece-Stremtan

, Marinelli

, Medicine

AoB

. ABM clinical protocol# 21: Guidelines for breastfeeding and substance use or substance use disorder, revised 2015. Breastfeed Med, 2015; 10(3):135–141.

36.

Regan

, Pereira

. Patterns of combustible and electronic cigarette use during pregnancy and associated pregnancy outcomes. Sci Reports, 2021; 11(1):1–9.

Perinatal Smoking and E-cigarette Use and Their Relationship with Breastfeeding: PRAMS 2015–2020