What Do Mothers Think about Concurrent Breastfeeding and Smoking?

Abstract

Background

According to newer AAP policies, smoking is not contraindicated with breastfeeding, yet smokers initiate and maintain breastfeeding less than non-smokers.

Objectives

1) Describe maternal knowledge and 2) attitudes regarding concurrent breastfeeding and smoking or nicotine replacement therapy (NRT) and 3) evaluate the association between maternal smoking and infant feeding practices.

Methods

Mothers bringing children <13 months old for an appointment completed a 24-item, anonymous survey which addressed knowledge, attitudes and practices about concurrent breastfeeding and smoking/NRT.

Results

Among 204 survey completers, 63% were African American, 52% had never breastfed and 54% had never smoked. Knowledge: Regardless of smoking status, 19% were aware of the recommendation to smoke after breastfeeding; most did not know that nicotine gum (42%) or patch (40%) transfers less or about the same amount of nicotine into breast milk than smoking a pack per day. Attitudes: Most mothers (80%) believe that women should not smoke any cigarettes if breastfeeding; current smokers (25%) were more likely than former (10%) or never smokers (11%) to find it acceptable to smoke one or more cigarettes per day (p=.03). Only 2% found it acceptable to use NRT while breastfeeding. Practice: Among ever breastfeeders, 10% stopped breastfeeding because of smoking. Over half of recent or current smokers reported that smoking impacted their infant feeding decision.

Conclusions

Mothers in this sample believe that women who smoke or take NRT should not breastfeed. Smoking status impacted women’s infant feeding practices. Correction of misinformation could increase breastfeeding rates.

BACKGROUND

The adverse consequences of maternal smoking and the benefits of breastfeeding on infant health are well documented. Recommendations about breastfeeding in mothers who smoke have evolved as new data on the relative risk-benefit relationship becomes available (1–3). In 2001, nicotine was removed from the American Academy of Pediatrics (AAP) list of drugs contraindicated during breastfeeding (4). Both the AAP and U.S. Department of Health and Human Services support breastfeeding among women who can’t or won’t stop smoking, because the benefits of breast milk outweigh the risks from nicotine exposure (4–6).

While supporting breastfeeding in the context of maternal smoking, the policies strongly encourage smoking cessation or reduction (4, 5). Limited evidence supports the use of nicotine replacement therapy (NRT) during pregnancy (7) and lactation (8) as a harm reduction strategy although no formal guidelines exist. When used as directed, the 21-mg nicotine patch transfers no more nicotine into breast milk than one pack-a-day smoking while 14- and 7-mg patches and nicotine gum confer less (8, 9). In order to minimize transfer of nicotine into breast milk, women are encouraged to remove the patch at bedtime and to refrain from breastfeeding for 2 to 3 hours after chewing the gum or smoking cigarettes (9). Smokers support the use of NRT during pregnancy (10) but their attitudes about NRT during lactation have not been reported.

Smokers are less likely to initiate and continue breastfeeding compared to nonsmokers (11–15). Physiologic factors don’t explain these differences (16). Donath suggested the difference is largely explained by lower rates of breastfeeding intention among smokers (17). Other possible explanations are that smokers are more likely to perceive their milk supply as insufficient (18), are less health conscious than the general population or may have concerns about adverse health effects of smoking on their baby.

The study objectives were to: 1) describe maternal knowledge and 2) attitudes regarding concurrent breastfeeding and smoking/NRT, and 3) evaluate the association between maternal smoking status and infant feeding practices.

METHODS

This cross-sectional study was conducted in a pediatric clinic associated with a large academic medical center and a pediatric resident continuity clinic. Anonymous surveys were collected from September 2005 to October 2006. During that time the clinic population was mostly African American (70%) and Caucasian (25%), low-income and primarily publicly insured (81% Medicaid) and had more than 18,000 annual visits. The University of Pittsburgh Institutional Review Board granted exempt study approval.

Population

Mothers older than 17 years were asked to complete the 24-item self-administered, multiple-choice survey if they brought a child less than 13 months old for a visit. Mothers were approached to participate in one of three ways: by registration staff, by the nurse rooming the child or by the physician seeing the patient. Mothers placed completed surveys into a slotted box at the nurse’s station or registration desk. Due to the distribution method, we are unable to determine a refusal rate. In order to minimize recall bias regarding infant feeding practices, mothers who delivered in the past year were targeted.

Survey Development

The medical literature was reviewed for previously validated questions related to smoking and breastfeeding; no prior surveys were identified. Therefore the authors developed questions based upon study aims. Modifications were made after review by experts in survey methods, tobacco research, breastfeeding medicine and general pediatrics. The survey was piloted among community women and feedback was reviewed with a survey methodologist and again modified.

Survey Items

Knowledge

Possible responses to, “Do you think it is better for the baby if a mother smokes before or after she breastfeeds her baby?” were: a) before or b) after breastfeeding, c) it doesn’t matter and d) I don’t know with b the correct response. In two separate items about NRT (patch and gum), the question read, “Do you think that nicotine gum/patch lets more, less or about the same amount of nicotine get into breast milk than smoking a pack of cigarettes per day?” Correct responses were “less” for gum and “same” or “less” for patch.

Attitude

Possible responses to, “In your opinion, what is the maximum number of cigarettes a day a woman can smoke and still breastfeed her child?” were zero, 1–5, 6–10, 11–15, 16–20, more than 20. Possible response were yes, no or don’t know to, “Do you think it is okay for the baby if a woman uses nicotine gum or the nicotine patch while she breastfeeds her baby?”

Smoking and infant feeding practices

All respondents were asked, “If you ever breastfed your baby, mark the top 3 reasons why you stopped breastfeeding.” “Started smoking” was one of 14 possible responses. Women who smoked in the past year were asked, “Did smoking ever affect your decision whether or not to breastfeed your baby?” and possible positive responses included “I decided not to smoke while breastfeeding”, “I decided to smoke fewer cigarettes while breastfeeding” and “I decided not to breastfeed because I smoked.” Women could also respond that smoking did not affect their decision.

Smoking Status

Women selected from the following smoking descriptions: a) I never smoked, b) I smoked fewer than 100 cigarettes in my lifetime, c) I stopped smoking over a year ago, d) I stopped smoking less than a year ago and e) I currently smoke. Women who responded a or b were categorized as non-smokers (19); c or d as former smokers; d recent quitters and e current smokers.

Breastfeeding Status

Women who ever breastfed were asked how long they breastfed their youngest child. Responses were grouped as never breastfed their youngest baby and breastfed <1 month, 1 to 4 months, longer than 4 months, and still breastfeeding. For women who were still breastfeeding, duration of breastfeeding was defined as the infant’s age at the time of survey completion, thus underestimating breastfeeding duration for this subgroup of women.

Analysis

Responses to maternal knowledge and attitudes were compared across smoking status, breastfeeding status, maternal race and infant age at survey completion (birth-4 months and >4 months) with Pearson χ² and Fisher’s exact tests (when expected cell sizes <10). When categorical variables had an underlying order, exact Mann-Whitney U statistics were used to account for ordering. Analyses were repeated using two additional smoking status definitions (1. not current vs. current smokers; 2. non-smokers plus former smokers stopped greater than one year ago vs. recent quitters plus current smokers). Because results did not change with these smoking status groupings, they are not reported. Statistical analyses were conducted using SPSS version 14.0 (SPSS, Inc Chicago, IL) and StatXact v.7 (Cytel Inc, Cambridge, MA).

RESULTS

Sample (Table 1)

Table 1.

Demographic characteristics of study population by smoking status; percent of each group

Characteristic		% with Characteristic
		All n=204	Non-Smoker n=110	Former Smoker n=39	Current Smoker n=55	P
Maternal Age (years)	18–21	32	34	26	33	0.6£
	22–25	25	25	20	28
	26–30	24	22	36	20
	>30	19	20	18	19
Race	White	33	21	56	38	<0.001†
	Black	63	69	44	60
	Other	4	9	0	2
Baby age (months)	< 1	39	39	41	37	0.7£
	1 to 4	26	30	19	22
	5 to 12	36	32	41	41
Ever breastfeed any baby?	Yes	48	49	58	40	0.2†
Duration breastfed youngest baby? (months)*	Never	54	53	43	62	0.3£
	<1	23	22	30	20
	1–4	14	16	16	9
	>4	10	9	11	9

^ƒChi-square

^†Fisher’s exact test

^£Kruskal Wallace

^*For women who were still breastfeeding (n=33) when they completed the survey, duration of breastfeeding was defined as age of the child at the time of survey completion, thus underestimating duration of breastfeeding among this population.

Non-smokers: mothers who never smoked or smoked fewer than 100 cigarettes in their lifetime.

Former smokers: mothers who smoked in the past but were no longer smoking at the time they completed the survey.

Current smokers: mothers who were smoking at the time they completed the survey.

Surveys were completed by 209 women; five (2%) did not indicate smoking status and were excluded from analyses. Among 204 complete responders, over half were African American, aged 18–25 years, had never breastfed any child and had never smoked. White women were more likely to be former smokers but were not different with respect to age and breastfeeding status. The proportion of women who breastfed their youngest child was similar across smoking status.

Objective 1: Knowledge

Regardless of smoking status, few women (19%) were aware of recommendation to smoke after breastfeeding. Mothers who breastfed their youngest baby were more likely to respond correctly than mothers who did not (26 vs. 11%, p=.02). Many mothers did not know that nicotine gum (42%) or patch (40%) transfers less or about the same amount of nicotine into breast milk than smoking a pack-per-day. Knowledge was not associated with maternal age, race or child’s age at survey completion.

Objective 2: Attitude (Table 2)

Table 2.

Mother’s attitudes and practices towards breastfeeding and smoking/NRT, percent of each group

Survey Item	Non-Smoker n=110	Former Smoker n=39	Current Smoker n=55	p†
In your opinion, what is the maximum cigarettes per day a woman can smoke and breastfeed her baby?
0	89	89	74	.03£
1 – 5	9	5	14
> 5	2	5	11
Do you think it is okay for the baby if a woman uses NRT while she breastfeeds her baby?
Yes	2	3	4	0.4†
No	70	82	64
Don’t know	28	15	33
Which of the following best describes what you would do if you were a mother who smokes cigarettes?
Keep smoking and breastfeed	4	0	31	<.001†
Stop smoking and breastfeed	74	70	33
Smoke and don’t breastfeed	23	30	36
For women who smoked in the past year, Did smoking ever affect your decision whether or not to breastfeed your baby?
	Recent Quitter		Current smoker
	n=12*		n=52*
Did not smoke and breastfed	67		15	.002†
Smoked fewer cigarettes and breastfed	0		12
Smoked and didn’t breastfeed	17		31
Smoking didn’t affect decision	17		42

^†Fisher’s exact test

^£Kruskal Wallace

^*5 recent quitters and 3 current smokers did not respond to this survey item

cpd: cigarettes per day

NRT: Nicotine replacement therapy (patch or gum)

Most respondents (85%) answered that women can’t smoke any cigarettes if breastfeeding and only two responded that women can smoke more than 10 cpd. Current smokers (25%) were more likely than former (10%) or never smokers (11%) to respond that breastfeeding mothers can smoke any cigarettes (p=.03). Most mothers (70%), regardless of smoking status, indicated that it was not acceptable to use NRT while breastfeeding. When asked what a woman who smokes should do if she wants to breastfeed, non-smokers were most likely to suggest she not smoke and breastfeed as opposed to not breastfeed and smoke. In contrast, current smokers were more evenly divided on what they would advise; a third each would suggest women breastfeed and smoke, breastfeed and stop smoking and not breastfeed and smoke. No recent quitters recommended that women breastfeed and smoke. Attitude towards smoking and breastfeeding did not differ by maternal race, age, breastfeeding experience or child’s age at survey completion.

Objective 3: Infant Feeding Practice and Smoking Status (Table 2)

Among ever breastfeeders, 10% (9/99) marked smoking as one reason they stopped breastfeeding. Among recent quitters (n=17) and current smokers, 66% reported that smoking affected their infant feeding practice; 18/40 didn’t breastfeed and 22/40 cut down on or didn’t smoke in order to breastfeed. Recent quitters were more likely to have quit smoking to breastfeed. Among women who didn’t smoke in order to breastfeed (n=16), half were smoking again (8/16) at survey completion.

DISCUSSION

This was the first study in which mothers were asked their opinions about concurrent breastfeeding and smoking or NRT. Women lack knowledge of and have negative attitudes towards concurrent breastfeeding and smoking or NRT. Despite little knowledge, among women who smoked in the past year, over half indicated that their smoking status influenced their infant feeding decision. The study findings suggest that women’s knowledge and attitudes about concurrent breastfeeding and smoking could, in part, explain why smokers are less likely to breastfeed than nonsmokers.

Current smokers are more accepting of breastfeeding mothers smoking low numbers of cigarettes than non-smokers. Possible reasons for this difference include smokers’ increased comfort with unhealthy behaviors or managing risk-benefit regarding smoking or increased awareness of recent policy changes. While smokers have some comfort with concurrent breastfeeding and smoking, few find it acceptable to breastfeed with NRT. This contrasts with Griffith’s survey findings that 68% of smokers in late pregnancy would accept NRT during pregnancy (10). Perhaps mothers are not aware that nicotine passes more readily across the placenta than into breast milk. The apparent difference in attitudes towards NRT between pregnant and postpartum women deserves further study.

In contrast to previously published studies, we found no difference in breastfeeding rates by smoking status. Liu found that non-smokers and former smokers breastfed at similar rates but that women who relapsed to smoking early after delivery or smoked throughout pregnancy were less likely to breastfeed (15). Our sample may be too small to detect a difference in breastfeeding rates or differences in our study population may explain the disparate findings. We studied primarily low-income, African American women. Liu studied mostly non-Hispanic white women with a high breastfeeding rate (91%). Our 48% breastfeeding rate is significantly below the 72% national rate and 61% African Americans rate (20).

Our findings may not generalize to all populations. The study was conducted at a single inner-city clinic that has on-site breastfeeding and smoking cessation support. Providers are educated to support breastfeeding regardless of smoking status and to offer NRT to breastfeeding women. This should bias the results towards higher acceptability of smoking and NRT with breastfeeding but few women found it acceptable even in our clinic. The survey did not allow for open ended responses thus forcing women to select among the attitude statements offered. Women up to one year post delivery were asked to consider their infant feeding decisions which could introduce recall bias. However, two thirds of mothers completed the survey within 4 months of delivery. Despite being an anonymous survey, women may not admit they smoked while breastfeeding; biological measures of smoking weren’t obtained.

Women’s knowledge, attitudes and practices may reflect a failure of the medical community to properly educate women during pregnancy and postpartum period about the relative risks and benefits of breastfeeding vs. formula feeding in the context of smoking and NRT.

Abbreviations

AAP

American Academy of Pediatrics

cpd

cigarettes per day

NRT

nicotine replacement therapy