Gestational Weight Gain Among Urban and Rural Pregnant Individuals Who Smoke or Quit Smoking

Abstract

Introduction:

Approximately 7.2% of individuals in the United States smoke during pregnancy, and smoking cessation is associated with excessive gestational weight gain (GWG). Weight gain is a common reason for not quitting smoking or relapsing. The current study aimed to characterize who is most at risk for excessive GWG and determine the moderating effect of rurality given the higher smoking rates and lower access to healthcare services in these areas.

Methods:

Data from the Virginia Pregnancy Risk Assessment Monitoring System (PRAMS; years 2009-2020) were used to assess the association between participant characteristics, smoking behaviors, and rurality by excessive GWG status.

Results:

Almost half (44.0%) of participants experienced excessive GWG; 9.8% of participants quit smoking while 6.9% continued smoking. Respondents who quit during pregnancy had higher odds of excessive GWG than non-smoking respondents (OR = 1.83, 95% CI: [1.24, 2.71]). Among those who were non-smoking, respondents in rural areas, compared to urban areas, had a higher probability of experiencing excessive GWG (0.46 vs. 0.44, p<0.001). For those who quit smoking (0.60 vs. 0.41, p<0.001) or continued to smoke during pregnancy (0.46 vs. 0.33, p<0.001), urban residence was associated with a higher likelihood of excessive GWG compared to rural residence.

Conclusions:

Smoking cessation and weight management during pregnancy are critical to promoting infant and maternal health. Targeted interventions combining weight management and smoking cessation have been successful among the general population and could be adapted for pregnant individuals who smoke to facilitate cessation and healthy GWG in both urban and rural areas.

Approximately 7.2% of individuals in the United States smoke during pregnancy,¹ incurring negative infant health effects.² Quitting smoking during pregnancy can lead to excessive gestational weight gain (GWG),^3,4 since smoking cessation is associated with weight gain.⁵ This weight gain is a commonly cited reason for not quitting smoking and relapse.^6–10 This is particularly relevant for pregnant people already experiencing weight gain, which can hinder their efforts to quit.¹¹ Thus, understanding facilitators for continued or quitting smoking during pregnancy can inform targeted interventions for smoking cessation, relapse prevention, and healthy GWG.

Pre-pregnancy overweight/obesity^4,12,13 and lower education levels,^12,14 are previously established risk factors for excessive GWG, while racially minoritized groups^12,13 are less likely to have excessive GWG. However, it is unclear if these associations are different in the context of smoking status. Furthermore, the association between rurality and excessive GWG is unknown among those who smoke or have quit. This is important given the decreased availability of healthcare services in rural areas. Given the disparate rates of smoking and subsequent health consequences among those with lower levels of education, who identify as Black,¹⁵ or live in rural areas,^16,17 it is crucial to understand who is most at risk, and the intersection of where individuals live with smoking behaviors and excessive GWG, in efforts to promote health equity.

This study examines GWG among individuals who either smoked or quit smoking during pregnancy and the moderation effect of rurality. We hypothesized that quitting, identifying as White, and having lower education levels would be associated with excessive GWG and individuals in rural areas would have lower risk for excessive GWG due to higher smoking rates during pregnancy in rural areas.¹⁷

Methods

Study Sample

Data were from the Virginia Pregnancy Risk Assessment Monitoring System (PRAMS) collected between 2009 to 2020.¹⁸ Participants included 5,993 adults (>18 years old) from Virginia, excluding those without data on GWG (N = 979) and with pre-term delivery (<37 weeks of gestation; N = 1,481) since categorizing their GWG was not feasible.¹⁹ Study procedures were approved by the Virginia Department of Health and the University of Virginia.

Measures

Participants self-reported race, ethnicity, education, and age, and were asked about their smoking behaviors three months before pregnancy and during the last three months of pregnancy. Participants were categorized as: non-smoking (not smoking during either period), quit during pregnancy (smoked three months before pregnancy but not the last trimester), and smoking during pregnancy (smoked three months before pregnancy and the last trimester). One participant denied smoking three months prior to pregnancy and endorsed smoking in the last trimester, and was excluded from analysis given the small sample and our inability to determine if this was initial uptake or a relapse from a prior quit.

Self-reported pre-pregnancy height and weight were used to calculate pre-pregnancy BMI using the standard categorical cut points.²⁰ To establish GWG, participants were asked to report amount of weight gained or lost during pregnancy. Participants were categorized with excessive GWG based on the National Academy of Medicine guidelines (normal weight: >16 kg, overweight: >11.5 kg, obesity: >9 kg).¹⁹ Lastly, counties were classified using the 2013 rural–urban continuum codes (1-3: “urban”, 4-9: “rural”).²¹

Statistical Analysis

Logistic regression was used to examine the association between participant characteristics, smoking behaviors, and rurality by excessive GWG status. Moderation analysis was conducted using the interaction between rurality and smoking status. Models controlled for participant characteristics, and clustered robust standard errors were used. All analyses were performed using STATA 18 in 2023.

Results

Almost half (44.0%) of participants experienced excessive GWG and 9.8% quit smoking, 6.9% continued smoking, and 83.3% were non-smoking (Table 1). Table 2 presents an adjusted model predicting excessive GWG based on participant characteristics. Rural participants had higher rates of smoking during pregnancy than urban participants (16.1% versus 6.2%, respectively). Respondents >30 years old had lower odds of excessive GWG compared to those <24 years. Having a college or higher education, urban residence, and having overweight or obesity prior to pregnancy were associated with increased odds of excessive GWG, while identifying as a racial minority was associated with lower odds. Respondents who quit smoking during pregnancy also had higher odds of excessive GWG than those who were non-smoking.

Table 1.

Summary Statistics of the Sample by Excessive Gestational Weight Gain (GWG) Status

	Did not have excessive GWG N (%)	Excessive GWG N (%)	Total N (%)	p
Total	3,356 (56.0%)	2,637 (44.0%)	5,993 (100.0%)
Age categories (years)				0.02
18-24	535 (16.0%)	469 (17.8%)	1,004 (16.8%)
25-29	928 (27.7%)	787 (29.9%)	1,715 (28.7%)
30-34	1,123 (33.5%)	837 (31.8%)	1,960 (32.7%)
35+	764 (22.8%)	543 (20.6%)	1,307 (21.8%)
Education				0.006
Some college or less	1,708 (50.9%)	1,437 (54.5%)	3,145 (52.5%)
College degree or more	1,648 (49.1%)	1,200 (45.5%)	2,848 (47.5%)
Race				0.08
Racial minority	1,130 (33.7%)	831 (31.6%)	1,961 (32.8%)
White	2,221 (66.3%)	1,801 (68.4%)	4,022 (67.2%)
Ethnicity				0.48
Hispanic	329 (9.8%)	244 (9.3%)	573 (9.6%)
Non-Hispanic	3,025 (90.2%)	2,390 (90.7%)	5,415 (90.4%)
Rurality				0.60
Rural	250 (7.4%)	206 (7.8%)	456 (7.6%)
Urban	3,106 (92.6%)	2,431 (92.2%)	5,537 (92.4%)
Pre-Pregnancy Body Mass Index				<0.001
Underweight	217 (6.5%)	0 (0.0%)	217 (3.6%)
Normal weight	2,078 (61.9%)	956 (36.3%)	3,034 (50.6%)
Overweight	540 (16.1%)	881 (33.4%)	1,421 (23.7%)
Obesity	521 (15.5%)	800 (30.3%)	1,321 (22.0%)
Smoking status during pregnancy				<0.001
Non-smoking	2,837 (85.7%)	2,080 (80.2%)	4,917 (83.3%)
Quit smoking during pregnancy	239 (7.2%)	340 (13.1%)	579 (9.8%)
Smoked during pregnancy	236 (7.1%)	172 (6.6%)	408 (6.9%)

Note. GWG = gestational weight gain. Values within categories that do not add up to the total N reflect missing data within the respective variables.

Table 2.

Adjusted Odds Ratio Estimation of Excessive GWG on Rurality and Respondent’s Characteristics

Dependent variable: Excessive GWG (binary variable)
	Main Model	Moderation Analysis
	Odds Ratio [95% CI]	Odds Ratio [95% CI]
Age categories (Ref. 18-24 years)
25-29 years	0.94 [0.83,1.07]	0.93 [0.80,1.09]
30-34 years	0.81*** [0.75,0.88]	0.81*** [0.74,0.89]
35+ years	0.75*** [0.65,0.86]	0.75*** [0.64,0.87]
Education (Ref. Some college or less)
College or more	1.14*** [1.04,1.26]	1.15*** [1.06,1.24]
Race (Ref. White)
Racial minority	0.77*** [0.76,0.78]	0.77*** [0.76,0.77]
Ethnicity (Ref. Non-Hispanic)
Hispanic	0.94 [0.87,1.01]	0.94* [0.88,1.01]
Rurality (Ref. rural)
Urban	1.16*** [1.14,1.19]	0.92*** [0.90,0.94]
Smoking status during pregnancy (Ref. non-smoking)
Quit smoking in pregnancy	1.83*** [1.24,2.71]	0.79*** [0.75,0.83]
Smoked during pregnancy	0.99 [0.67,1.46]	0.54*** [0.51,0.57]
Pre-pregnancy BMI (Ref. Normal weight)
Overweight	3.60*** [3.39,3.81]	3.61*** [3.36,3.89]
Obesity	3.50*** [3.49,3.50]	3.50*** [3.48,3.51]
Smoking status × Rurality (Ref. non-smoking × Rural)
Quit smoking in pregnancy × Urban		2.56*** [2.53,2.59]
Smoked during pregnancy × Urban		2.02*** [2.02,2.03]
Observations	5670	5670

Note.

^*p < .05;

^**p < .01;

^***p < .001.

GWG = gestational weight gain; BMI = body mass index; CI = confidence interval.

Table 2 and Figure 1 show the results from the moderation analyses to investigate the interaction between rurality and smoking status on excessive GWG. Furthermore, Table 3 displays marginal effects estimates. Among those who were non-smoking, rural respondents had a similar, but statistically significant probability of experiencing excessive GWG than urban respondents (0.46 vs. 0.44, p<0.001). However, for those who quit smoking (0.60 vs. 0.41, p<0.001) or continued to smoke (0.46 vs. 0.33, p<0.001), urban residence was associated with a higher likelihood of excessive GWG than rural residence.

Figure 1:

Moderation effect of rurality and smoking status on excessive GWG among respondents.

Note. GWG = gestational weight gain.

Table 3.

Marginal Effects Estimation Based on Moderation Analysis

Smoking Status × Rurality	Marginal effect	Standard Error	p
Non-smoking × Rural	0.46	0.002	<0.001
Non-smoking × Urban	0.44	0.001	<0.001
Quit smoking × Rural	0.41	0.007	<0.001
Quit smoking × Urban	0.60	0.004	<0.001
Smoked during pregnancy × Rural	0.33	0.006	<0.001
Smoked during pregnancy × Urban	0.46	0.005	<0.001

Discussion

Participants who quit smoking had an 83% higher likelihood of experiencing excessive GWG, consistent with previous studies,^22–27 emphasizing the importance of combining weight management and smoking cessation interventions for this high-risk population. Our novel findings indicate rurality increased the likelihood of excessive GWG in the absence of smoking, while urban residence was associated with higher risk for those who quit or continued smoking. Thus, the relationship between excessive GWG and rurality is more nuanced than previously thought,²⁸ and individuals in urban areas may need additional resources to help combat excessive GWG in the context of smoking while individuals in rural areas may need excessive GWG interventions more globally. Additionally, 6.9% of respondents continued to smoke in the last trimester, which highlights the need for smoking cessation interventions during pregnancy in Virginia.

Contrary to previous findings,^4,29 respondents with higher education were more likely to experience excessive GWG; these findings may be attributable to the influence of smoking status, as lower education levels are robustly linked to higher smoking rates.¹⁵ Thus, individuals with lower education levels are smoking at higher rates, which may be preventing excessive GWG. Additionally, respondents identifying as racial minorities were less likely to experience excessive GWG (controlling for individual characteristics), consistent with other general population studies.¹³

Limitations

The study has some limitations. Most participants were from urban areas; thus, future research should include larger samples of rural participants. The study also did not consider the number of cigarettes smoked or pregnant individuals who replaced cigarettes with e-cigarettes, warranting exploration of the potential impact of e-cigarette use on GWG. The study also relied on self-reported weight data, which may be impacted by recall bias or reluctance to accurately share sensitive information. Lastly, we only assessed excessive GWG. Future studies may benefit from assessing inadequate GWG in the context of smoking.

Conclusions

Smoking cessation and weight management are critical to promoting infant and maternal health. Targeted interventions combining weight management and smoking cessation, successful among the general population,³⁰ can be adapted for pregnant smokers, particularly in medically underserved regions (e.g., rural areas). While our study shows higher excessive GWG in urban areas, it is essential to consider elevated smoking rates in rural areas. Therefore, addressing smoking cessation and healthy GWG in both urban and rural areas is crucial for healthy pregnancy outcomes.