Sage Journals: Discover world-class research

Abstract

Introduction

It is unclear why people of Mexican ancestry who smoke report relatively high rates of discussing quitting-related topics, which predicts cessation attempts.

Methods

Using Ecological Momentary Assessment (EMA), adults from Mexico (n=40) and with Mexican heritage living in the US (n=52) who smoked daily were sent texts/emails each morning with cessation tips and cessation benefits over 15 days. At baseline, participants reported their personal network characteristics across up to eight different alters. Each evening of the 15 days, participants reported forgoing any cigarettes they normally would smoke (yes/no) and having had any quitting-related conversations (yes/no). Multilevel logistic models were estimated to regress these day-level outcomes of forgoing and, separately, quitting-related conversation on the number of EMA surveys participants had completed prior to that day, network characteristics, and baseline sociodemographic and smoking-related covariates.

Results

Participants were more likely to forgo cigarettes if they reported a higher proportion of female alters (AOR = 2.20, 95% CI 1.16–4.18), had more quitter alters whose opinions they respected (AOR = 6.14, 95% CI 1.41–26.75), and as the number of prior EMA surveys completed increased (AOR = 1.19, 95% CI 1.14–1.25). Mexican Americans who preferred English were significantly less likely to forgo smoking (AOR = 0.20, 95% CI 0.04–0.99) compared to participants residing in Mexico. Greater social bonding (i.e., tightly knit networks) was positively associated with having quitting-related conversations (AOR = 1.45, 95% CI 1.05–1.99).

Conclusions

Combining targeted messages with activation of quit-support within networks may influence intermediary behaviors that facilitate smoking cessation among Latino/a populations, including individuals not intending to quit soon.

Keywords

smoking cessation tobacco use digital health intervention efficacy messages social impact theory

Introduction

Latino/as, the largest racial minority in the United States (US; 19.5% of the population, with two-thirds of Mexican descent),¹ attempt to quit smoking more often than non-Latino/a White smokers but face greater barriers in receiving advice to quit or using evidence-based cessation methods and treatments.^2-4 These disparities are further stratified by socioeconomic status (SES), with lower-SES individuals experiencing greater challenges in cessation success⁵ and being more likely to have social networks composed of other lower-SES individuals.⁶ This paper’s main aim is to provide initial evidence of the role of social networks in the context of a low-cost intervention designed to encourage Latino/as from low educational background to quit smoking through short messages that highlight the benefits of quitting and provide tips to support cessation (hereafter, ‘efficacy messages’), sent daily to respondents via text or email over two weeks.

A substantial body of research shows that network structure influences both health behaviors, including smoking,^7,8 and the cognitive elaboration of health information, including smoking-related behaviors.⁹ Indeed, smoking cessation and continued smoking in adulthood are strongly associated with the composition and structure of individuals’ social networks.¹⁰ Individuals surrounded by more smokers are less likely to quit and more prone to relapse.^10-13 Moreover, smoking behavior tends to cluster within social networks, and interpersonal influence in these networks may help explain similarities in smoking patterns among socially connected individuals.¹⁰

Social network theory provides a robust framework for conceptualizing how behaviors and attitudes diffuse through interpersonal connections.¹⁴ (Dynamic) Social Impact Theory,^15,16 supported by extensive basic and applied research,¹⁷ posits that social influence on an individual is a multiplicative function of strength (i.e., the perceived authority or importance of the influence source); immediacy (i.e., the spatial or temporal proximity of the influence source); and number¹ (i.e., the size of the influencing group). Building on this framework, we develop a weighted version of the E–I index,¹⁸ which has long been used to study segregation,^19,20 to account for the prevalence of quitters (vis-à-vis non-quitters) in an individual’s network (i.e., in Dynamic Social Impact Theory's parlance, these groups are the basis of 'persuasive social impact' [influence by change-supporting alters] and 'supportive social impact' [influence by change-opposing alters], respectively)¹⁶ weighted by the immediacy and strength of their respective influence or social impact. This adaptation enables a nuanced analysis of social influence processes relevant to smoking cessation.

Health communication literature suggests that interpersonal communication acts as a network-level mechanism, amplifying the effects of health messages by facilitating their dissemination and reinforcing their influence within social groups.^8,21,22 Conversations about health messages may promote cognitive elaboration, encouraging individuals to critically assess and engage with messages (e.g., anti-smoking messages),⁹ thereby increasing the likelihood of behavioral changes, such as attempting to quit smoking.²³ This mechanism may be especially relevant for Latino/as, who, when compared to non-Latino/a Whites, report more frequent interpersonal discussions about anti-smoking messages and health warning labels.^24,25 Moreover, these conversations have been shown to predict cessation attempts across different ethnic groups, suggesting a cultural gradient wherein stronger identification with Latino/a or Mexican heritage and Spanish language use may further promote message sharing and engagement.

Despite this evidence, existing studies have concentrated on messages that emphasize the negative consequences of smoking in anti-smoking communication campaigns^22,26 and warning labels on cigarette packs.^24,25,27,28 Yet, evidence shows that more positive ‘efficacy messages’ can be effective in their own right given that, when included inside cigarette packs as elaborated messages, they increase cessation-related cognitions and behaviors, such as forgoing cigarettes.^29-33

Methodologically, much of the existing research on interpersonal communication and smoking relies on retrospective or experimental data, which are limited by recall bias and low ecological validity, respectively.³⁴ Ecological Momentary Assessment (EMA) overcomes these issues by collecting real-time self-reports near actual behaviors, enhancing validity and providing rich longitudinal data on experiences and behaviors.³⁵ This approach enables us to examine how conversations within personal networks may influence the early steps of smoking cessation in everyday life. This focus is especially important since closer, real-time measurements can better capture cognition-behavior relationships.³⁶ The present exploratory study utilizes novel EMA data, where participants were exposed to efficacy messages each morning for 15 consecutive days and completed a nightly survey about their smoking-related behaviors over the prior 24 hours. The sample includes individuals of Mexican origin residing in both Mexico and the United States. This study may generate preliminary evidence on short-term behavioral mechanisms and social network influence processes that could inform the development of future network-based interventions leveraging key precursors to smoking cessation among Latino/as.

Methods

Data Source

We analyzed data from an EMA study conducted between March 2022 and October 2023, involving Mexicans in Mexico (n=40) and Mexican Americans in the U.S. (n=52) who smoked daily. For 15 days, participants received a morning efficacy message and completed an evening survey on smoking and conversations. Recruitment combined social media, online panels, community outreach, and referrals. Eligibility required self-identification as Mexican/Mexican American, age ≥21, high school diploma or less, and smartphone ownership. Data were collected in 25 cohorts (1–18 participants). We excluded isolates (n=4) and cases with incomplete alter data (n=13), yielding 1,283 EMA observation days from 92 individuals with complete network and behavioral information.

Study Protocol

This study employed an exploratory EMA protocol to evaluate short-term responses to daily efficacy messages and examine how social network characteristics relate to cessation-related behaviors. At baseline, participants completed online measures (Qualtrics), viewed a brief training video, and took a comprehension quiz. During the 15-day study (including two weekends), individuals received one of 14 efficacy messages, adapted from prior research,^29-32 which were randomly selected (without replacement) and delivered at 8:30 a.m. in the respondents’ respective time zones via text or email in English or Spanish. In other words, all participants were potentially exposed to the same set of 14 messages, with each cohort following a different sequence. Messages combined text and graphics, highlighting cessation benefits and tips of cessation (Appendix A). Prior work shows such messages influence cessation cognitions independently of whether they are also exposed to “fear appeal” messages about the harms of smoking,^32,33 which are commonly mandated for cigarette package warning labels. Additionally, participants were prompted to complete a brief nightly EMA survey at 8:00 p.m. in their respective time zones. The nightly EMA survey, which took approximately 3 to 5 minutes to complete, collected information on participants’ smoking behaviors and interpersonal conversations related to smoking during the preceding 24 hours.

Compliance and Ethics

To maximize compliance, participants received nightly reminders at 10:00 p.m. and follow-up reminders the next day for any missed surveys. Staff conducted check-ins on days 1, 3, and 7, with follow-ups after two missed surveys. Participants were compensated with electronic gift cards in proportion to the surveys they completed. All procedures were approved by the University of South Carolina Institutional Review Board, and informed consent was obtained from all participants prior to their participation.

Measures

EMA Cessation-Related Behaviors

Each evening, participants reported whether they had forgone any cigarettes they would typically have smoked (yes/no) and whether they had engaged in a quit-related conversation within the prior 24 hours. If participants indicated that they had discussed quitting smoking with a network member on the survey day, they were asked to specify the content of the conversation, including whether they discussed strategies to quit smoking, received encouragement from the network member to quit smoking, expressed their own desire to quit smoking, or talked about the benefits of quitting smoking. Any such report was coded as a quit-related conversation (yes/no). Participation was tracked with a time-varying “EMA day” variable (i.e., cumulative number of prior daily surveys completed at the time of the survey, or time t). Each participant contributed up to 15 observation-days; on average, respondents completed 13.9 (SD = 1.4), indicating high compliance.

Social Network Measurements

At baseline, following Pachucki and Leal,³⁷ participants identified up to eight alters (aged 16+) using two name generators: (1) those “with whom you spent the most free time” and (2) those “with whom you felt close discussing important matters.” The name generators were shown in random order. For each alter, participants reported smoking attributes and relational characteristics that were used to construct indices of ‘social bonding’ and ‘social bridging.’

Following the approach of Perry et al,³⁸ social bonding was defined as the degree of social integration and emotional support within a participant’s network. The bonding index was calculated as the sum of four components: (1) number of close ties (very/extremely close), (2) average frequency of contact, (3) proportion of kin ties, and (4) whether a spouse/partner was named. Additionally, following Perry et al, social bridging reflects respondents’ access to diverse and non-redundant networks.³⁸ The bridging index summed seven measures: (1) proportion of non-kin ties, (2) network size (0–8 alters), (3) sole-bridge indicator (ego linking at least two disconnected alters [yes/no]), (4) network diversity (distinct alter role types [e.g., friend, neighbor]), (5) number of weak ties (not very/extremely close), (6) effective size,³⁹ and (7) inverse ego network density.

To measure the social impact of quitters, we used a two-step approach. First, we applied the classic E–I index^18-20 to capture the relative number of quitters (former smokers) versus non-quitters (current and never smokers) in the respondents’ networks. We added 1, so the index ranges from 0 (all alters are non-quitters) to 2 (all alters are quitters), with 1 indicating balance. This transformation centers the index at a positive value and ensures all values are non-negative, as negative values would make interpretation less straightforward:

{E ‐ I}_{i_u n w e i g h t e d} = 1 + (\frac{{# q u i t t e r a l t e r s}_{i} - {# n o n q u i t t e r a l t e r s}_{i}}{{# q u i t t e r a l t e r s}_{i} + {# n o n q u i t t e r a l t e r s}_{i}})

(1)

Second, we weighted the classic E–I index by immediacy (spatial proximity, six levels) and strength (respect for alters’ opinions, five levels), per Social Impact Theory.¹⁵ Specifically, each ego–alter tie weight was:

w_{i j} = (\frac{{i m m e d i a c y}_{i j}}{6}) \times (\frac{{s t r e n g t h}_{i j}}{5})

(2)

The weighted counts of quitters and non-quitters were then defined as:

Q_{i} = \sum_{j = 1}^{# a l t e r s} w_{i j} q_{j}

(3)

N_{i} = \sum_{j = 1}^{# a l t e r s} w_{i j} ({1 - q}_{j})

(4)

where

q_{j} = 1

if alter

j

is a quitter, 0 otherwise. Substituting Eq. 3 and Eq. 4 into Eq. 1 yields the weighted E-I index:

Social impact of quitters or {E ‐ I}_{i_w e i g h t e d} = 1 + (\frac{Q_{i} - N_{i}}{Q_{i} + N_{i}})

(5)

Eq. 5 implements the E–I index weighted by strength and immediacy, consistent with Social Impact Theory, and is defined where $Q_{i} + N_{i} > 0$ .

In addition to the previous measures, for each identified alter, participants reported their alters’ sex (male, female), current smoking status (1 = smoker vs. 0 = no and don’t know), and their approval of ego’s own smoking (1 = strongly disapproves and disapproves vs. 0 = neither approves nor disapproves, approves, strongly approves, and don’t know).

Other Measurements

Baseline measurements included standard smoking-related variables: daily smoking frequency (1–5, 6–10, 10+ cigarettes) and quit intention (yes = plan to quit within 6 months; no = later, not planning, or don’t know).⁴⁰ Sociodemographics included sex (male or female) and age (in years). Using information about participants’ country (Mexico or the U.S.) and the language (Spanish or English), we derived three groups: Mexicans in Mexico, Spanish-preferring Mexican Americans, and English-preferring Mexican Americans.

Analysis

We described baseline characteristics and daily outcomes across 1,283 observation days from 92 participants. Daily outcomes included the number of previous surveys the participant had completed before the EMA nightly survey that provided the observation of interest, forwent smoking at least one cigarette, and engaged in at least one quit-related conversation. Predictors of forgoing and engaging in conversations about quitting within the preceding 24 hours were estimated using multilevel mixed-effects logistic regression, reporting crude and adjusted odds ratios (OR and AOR, respectively). We estimated models with random intercepts for each respondent to account for repeated daily observations. Covariates included EMA day (i.e., prior daily surveys completed) and network measures (i.e., the number of females in the network, the number of alters who disapprove smoking, the number of smokers, the social impact of quitters, social bridging, and social bonding). Sensitivity analyses tested robustness. All analyses were conducted in Stata 18.5 (StataCorp, College Station, TX).

Results

The analytic sample consisted of Mexicans living in Mexico (43.5%) and Mexican Americans in the US who preferred speaking Spanish (25.0%) or English (31.5%). The majority were male (67.4%), and the average age was 42.9 years (SD = 11.3). Most reported no intention to quit smoking (69.6%) and typically smoked between 6 and 10 cigarettes per day (42.4%, Table 1). On average, participants forwent cigarettes on 8.1 out of the 15 study days (SD = 5.1). Additionally, on average, respondents reported discussing quitting smoking 2.9 days (SD = 4.0) during the study period (Table 1).

Table 1.

Baseline Characteristics and Key Variables in EMA Study Participants (n = 92)

Variable	Count or mean	% or SD
Baseline characteristics
Sex (count, %)
Male	62	67.4
Female	30	32.6
Age (mean, sd [range])	42.9	11.3 [21-69]
Group (count, %)
Mexico	40	43.5
US Spanish	23	25.0
US English	29	31.5
Daily smoking frequency (count, %)
1- 5 cigs	18	19.6
6-10 cigs	39	42.4
>10 cigs	35	38.0
Intention to quit smoking (count, %)
No	64	69.6
Yes	28	30.4
Network Characteristics
Network females (mean, sd)	1.8	1.4
Network females (count, %)
None	14	15.2
One	27	29.4
2 or more	51	55.4
Smoking disapproving (mean, sd)	1.7	1.6
Network smokers (mean, sd)	1.7	1.4
Quitter’s social impact (mean, sd)	0.3	0.4
Social bridging (mean, sd)	13.8	3.8
Social bonding (mean, sd)	7.6	2.4
Daily measurements^a
Days with EMA evening surveys (mean, sd)	13.9	1.4
Days forgoing cigarettes (mean, sd)	8.1	5.1
Days with conversations about quitting (mean, sd)	2.9	4.0

^aEstimates are based on 1,283 observations collected from the 92 participants over 15 days of observation. Cigs = cigarettes. EMA = Ecological Momentary Assessment.

Factors Associated With Forgoing Cigarettes and Conversations About Quitting Smoking

Mexican Americans who preferred English were less likely to forgo smoking (AOR = 0.20, 95% CI [0.04–0.99]) than participants in Mexico. Increasing levels of participation in the study (i.e., ‘EMA day’) were associated with higher odds of forgoing cigarettes (AOR = 1.19, 95% CI [1.14–1.25], Table 2). The results remain consistent when a sensitivity analysis employs a daily indicator (i.e., EMA day dummy variable), suggesting a dose–response relationship (see Appendix B).

Table 2.

Multilevel Logistic Regression Analysis of Factors Associated With Forgoing Smoking

Variable	OR^a	(CI 95%)		p-value	AOR^a,b	(CI 95%)		p-value
Sex
Male	Ref.				Ref.
Female	2.94	(0.91 -	9.50)	0.071	1.81	(0.43 -	7.56)	0.418
Age	1.02	(0.97 -	1.08)	0.346	1.02	(0.96 -	1.08)	0.604
Group
Mexico	Ref.				Ref.
US Spanish	1.87	(0.47 -	7.44)	0.372	1.57	(0.28 -	8.64)	0.606
US English	0.38	(0.11 -	1.36)	0.137	0.20	(0.04 -	0.99)	0.049
Daily smoking frequency
1- 5 cigs	Ref.				Ref.
6-10 cigs	1.03	(0.22 -	4.72)	0.973	1.24	(0.23 -	6.56)	0.802
>10 cigs	2.08	(0.44 -	9.79)	0.355	2.89	(0.47 -	17.76)	0.253
Intention to quit smoking
No	Ref.				Ref.
Yes	2.85	(0.85 -	9.50)	0.089	3.22	(0.87 -	11.96)	0.080
EMA day	1.19	(1.14 -	1.25)	<0.001	1.19	(1.14 -	1.25)	<0.001
Network Characteristics
Network females	1.64	(1.08 -	2.47)	0.019	2.20	(1.16 -	4.18)	0.016
Smoking disapproving	0.95	(0.68 -	1.33)	0.756	0.64	(0.40 -	1.03)	0.068
Network smokers	1.13	(0.76 -	1.66)	0.550	1.49	(0.89 -	2.47)	0.127
Quitters’ social impact	4.66	(1.22 -	17.73)	0.024	6.14	(1.41 -	26.75)	0.016
Social bridging	1.03	(0.89 -	1.20)	0.662	0.84	(0.66 -	1.07)	0.161
Social bonding	1.18	(0.94 -	1.49)	0.152	1.09	(0.79 -	1.50)	0.619

Note. The analytical sample comprised 1,283 observations collected from 92 individuals. Statistically significant values are in bold. Cigs = cigarettes. EMA = Ecological Momentary Assessment.

^aOR (Odds Ratio) and AOR (Adjusted Odds Ratio).

^bAOR: Adjusted by all variables in the table.

Social network characteristics were significantly associated with a greater likelihood of forgoing cigarettes, including a higher number of female contacts within a participant’s network (AOR = 2.20, 95% CI [1.16–4.18]). Similarly, participants whose networks showed higher levels of quitter’s social impact (i.e., higher values of the E-I quitter index weighted by immediacy and strength) had significantly higher odds of forgoing cigarettes (AOR = 6.14, 95% CI [1.41–26.75], Table 2). Appendix D indicates that whether the E–I quitter index is entered as presented in Eq. 5, or entered by weighing it only by strength or immediacy —each in separate regressions controlling for all other variables —it is positively and significantly related to foregoing cigarettes. This suggests, as predicted by Social Impact Theory, that none of these three factors is solely responsible for the observed result. Importantly, Appendix C demonstrates that evidence of the social impact of quitters remains if it is measured as strength × immediacy × number of quitters, as originally proposed by Social Impact Theory.^15,16 We decided to use a more nuanced measure where number is based on the E-I index—instead of the simple number of quitters—because the E-I index captures a critical nuance: the relative number of quitters vs. non-quitters in the ego’s social network.

Finally, social bonding, defined as access to a tight-knit social network, was the only factor positively associated with a greater likelihood of engaging in conversations about quitting smoking (AOR = 1.45, 95% CI [1.05–1.99], Table 3).

Table 3.

Multilevel Logistic Regression Analysis of Factors Associated With Conversations About Quitting Smoking

Variable	OR^a	(CI 95%)		p-value	AOR^a,b	(CI 95%)		p-value
Sex
Male	Ref.				Ref.
Female	0.61	(0.18 -	2.06)	0.428	0.74	(0.17 -	3.28)	0.688
Age	1.02	(0.97 -	1.07)	0.419	1.02	(0.96 -	1.08)	0.592
Group
Mexico	Ref.				Ref.
US Spanish	0.54	(0.14 -	2.12)	0.374	0.66	(0.13 -	3.51)	0.629
US English	0.15	(0.04 -	0.58)	0.006	0.24	(0.05 -	1.25)	0.091
Daily smoking frequency
1- 5 cigs	Ref.				Ref.
6-10 cigs	0.77	(0.16 -	3.63)	0.739	0.81	(0.16 -	4.17)	0.805
>10 cigs	0.78	(0.16 -	3.78)	0.758	0.85	(0.15 -	4.95)	0.861
Intention to quit smoking
No	Ref.				Ref.
Yes	2.72	(0.79 -	9.29)	0.111	1.59	(0.45 -	5.57)	0.468
EMA day	1.00	(0.96 -	1.04)	0.954	1.00	(0.96 -	1.04)	0.948
Network Characteristics
Network females	1.52	(1.00 -	2.33)	0.053	1.43	(0.78 -	2.61)	0.248
Smoking disapproving	1.61	(1.13 -	2.28)	0.008	1.22	(0.77 -	1.95)	0.393
Network smokers	1.39	(0.94 -	2.06)	0.097	1.51	(0.92 -	2.48)	0.105
Quitter’s social impact	3.19	(0.86 -	11.82)	0.083	2.93	(0.72 -	11.87)	0.132
Social bridging	1.16	(1.00 -	1.35)	0.051	0.82	(0.64 -	1.05)	0.115
Social bonding	1.49	(1.16 -	1.91)	0.002	1.45	(1.05 -	1.99)	0.024

Note. The analytical sample comprised 1,283 observations collected from 92 individuals. Statistically significant values are in bold. Cigs = cigarettes. EMA = Ecological Momentary Assessment.

^aOR (Odds Ratio) and AOR (Adjusted Odds Ratio).

^bAOR: Adjusted by all variables in the table.

Discussion

This exploratory EMA study, grounded in communication and social network theory, shows that social network characteristics are associated with smoking-related behaviors among Mexican-American adults. Smokers with more female contacts and influential quitters in their networks were more likely to forgo cigarettes—a consistent predictor of smoking cessation attempts that is sensitive to exposure to anti-smoking messages on cigarette packs.^41,42 As participants completed more EMA nightly surveys, they were more likely to forgo cigarettes, possibly due to cumulative exposure to efficacy messages, though future research should confirm this. Overall, findings suggest that simple efficacy messages delivered via text or email can influence early behavioral precursors to smoking cessation among Latino/a smokers. Moreover, tight-knit networks—characterized by greater closeness, higher contact frequency, and stronger kinship—were associated with more conversations about quitting, which can help promote cessation attempts.^24,25

Online and smartphone-based delivery methods provide innovative, rapid, and cost-effective ways to disseminate evidence-based smoking cessation interventions.⁴³ In our study, participants were more likely to forgo cigarettes as the number of EMA assessments increased, suggesting a dose–response relationship. This pattern indicates that cumulative exposure to efficacy messages or repeated daily engagement may serve as a low-cost, scalable intervention. Similarly, a systematic review and meta-analysis of EMA-based interventions found positive short-term effects on smoking cessation, particularly among participants recruited from clinical settings, who were likely more motivated to quit.⁴³ In contrast, most participants in our study (70%) had no intention to quit within six months and were recruited from non-clinical settings, underscoring key differences from prior EMA research. Notably, intention to quit was not significantly correlated with forgoing smoking, suggesting that EMA-based interventions may encourage cessation-related behaviors even among smokers not planning to quit, which is the majority of smokers. Future research should advance our understanding of EMA-based interventions by exploring their potential for broader implementation among individuals not actively seeking to quit,^44,45 while concurrently establishing robust evidence on their long-term effectiveness in supporting the precursors of smoking cessation.

Networks with more female contacts were associated with a greater likelihood of forgoing cigarettes, suggesting that women in smokers’ social environments play a key role in shaping cessation behaviors, and healthy behaviors more generally. This may reflect the central roles women occupy in families and communities, often providing stronger social support than men.⁴⁶ Qualitative research among young adult Latino/as has shown that concern for family is a primary motivation to quit smoking. Some men even reported they would stop smoking if their partner became pregnant,⁴⁷ and interventions targeting expectant fathers have successfully promoted cessation.⁴⁸ These findings indicate that women may act as influential promoters of smoking cessation, though further research is needed to clarify the mechanisms of their influence and to develop strategies that strengthen the support they provide.⁴⁹

While exposure to smoking within social networks can undermine cessation efforts through peer influence or normalization of smoking, these same networks can also provide social support to quit.⁵⁰ In line with predictions from Social Impact Theory,¹⁵ higher levels of quitters’ social impact (i.e., a multiplicative function of quitters’ strength, immediacy, and number) were significantly associated with the likelihood of forgoing cigarettes. This is consistent with a systematic review and meta-analysis of randomized controlled trials, which found that peer-support interventions significantly increased smoking abstinence, particularly when the peer was a former smoker.⁵¹

Our study refines the understanding of nonsmokers’ roles in social networks by distinguishing former smokers who have successfully quit. This distinction highlights the potential for successful quitters to act as influential agents within smokers’ social networks, serving as catalysts for smoking cessation. Existing research has mainly highlighted the negative aspects of peer influence, such as how exposure to smokers in a person’s social environment can promote smoking behavior.^11,13,52 Much less attention has been given to the possible positive impact of network members who have quit. This gap suggests an underexplored opportunity to leverage these individuals in supporting cessation efforts among their close friends and family.

Our findings support Social Impact Theory as a useful framework for understanding how former smokers influence others within social networks. In addition to the number of sources (e.g., quitters), the theory posits that social influence is further determined by two factors: the strength or perceived importance of the quitter’s opinion to the focal individual (ego), and immediacy, referring to the physical (or temporal) proximity between the ego and the alter. These factors interact multiplicatively to produce social impact. In the literature, two primary pathways through which quitters promote cessation can be identified: first, by exerting subtle social pressure or reinforcing quitting as a prevailing social norm, and second, by providing informational, emotional, or instrumental support to those considering or attempting to quit.⁵³ Our results further suggest that the effectiveness of these mechanisms depends on the number of ties to quitters relative to smokers, the strength of those ties, and their physical closeness or immediacy. Future research should help elucidate the different pathways through which quitters affect the attitudes and behaviors of smokers within their networks, and how these social influences contribute to quit intentions and cessation, including the ways influence is exerted through different types of networks (e.g., interaction networks, sentiment networks, or role relation networks such as friendship networks).⁵⁴

Social bonding (i.e., having close, similar social connections) was positively associated with having quit-related discussions in our study. Such bonding likely creates a safe and trusting context for sharing personal experiences, including quitting strategies, success stories, and resources.^55,56 Evidence indicated that larger and more tightly knit social networks are linked to higher levels of emotional support and social influence among adult smokers attempting to quit.⁵⁶ Moreover, similar individuals (e.g., kin or alters who share life experiences with ego) are often more comfortable engaging in sensitive conversations,⁵⁷ such as those about quitting smoking. Additional studies have shown that smokers with supportive friends and family are more likely to intend to quit than those lacking such support.⁵⁸ Together, these findings suggest that social bonding not only builds trust but also helps normalize quit-related communication within a shared value system.

English-preferring Mexican Americans were less likely to forgo cigarettes (in adjusted models) and engage in quitting-related conversations (in non-adjusted models) compared with both Mexicans residing in Mexico and Spanish-preferring Mexican Americans. This pattern aligns with evidence that cultural orientation and language preference influence interpersonal communication prompted by anti-smoking messages.^24,25,59 Language preference often reflects levels of acculturation, with English-preferring Mexican Americans typically more acculturated to U.S. mainstream culture and less embedded in the collectivist communication norms common in Mexican and Spanish-speaking communities.⁶⁰ Research indicates that Mexican smokers, particularly those residing in Mexico, report higher levels of interpersonal communication about quitting smoking, possibly due to stronger community-oriented values.²⁵ Moreover, Spanish preference among Mexican-origin populations in the U.S. has been associated with a greater likelihood of discussing cigarette warning labels,²⁴ suggesting that language use functions not only as a marker of acculturation but also as an indicator of communication styles and social support structures. Further research should confirm this association, as the present study may have limited statistical power to draw firm conclusions.

While our study offers essential insights into smoking behaviors among Mexicans and Mexican Americans, several limitations must be acknowledged. First, social network characteristics were assessed only at baseline, potentially missing shifts in network dynamics over time, although we expect such shifts to be minimal given the brief two-week study period. Moreover, the observational design cannot definitively distinguish between peer influence, peer selection, and unmeasured confounding, the latter of which may affect observed associations and constrain the ability to draw strong causal inferences in studies of this type. Second, the modest sample size and focus on a specific Latino/a subgroup may limit both generalizability and statistical power to detect differences between subpopulations. This sample size is in the range used in many intensive EMA protocols, while also reflecting feasibility constraints around recruitment and the exploratory nature of this study, which aimed to provide initial evidence on mechanisms linking social network characteristics and cessation-related behaviors. Finally, the short study duration limits the ability to draw conclusions about long-term cessation outcomes and the role of quit-related conversations. However, the 15-day window was intentionally selected to capture short-term, real-time processes using EMA, which is best suited for examining immediate behavioral responses, moment-to-moment fluctuations in decision-making, and interpersonal influences. This early-phase design allows for precise assessment of proximal indicators of cessation—such as forgoing cigarettes and engaging in quit-related conversations—that are consistent predictors of subsequent cessation attempts, including among Mexicans and U.S. Latino/as. Moreover, the use of daily EMA reduces recall bias due to the shorter time frames on which people report behaviors, thereby yielding a more accurate approximation of those behaviors in “real-time” compared to typical surveys that require recall of events that take place over much longer periods of time. In this way, the study’s use of intensive data collection over a relatively short duration aligns with its focus on early precursors of cessation rather than on long-term abstinence trajectories that would require larger samples and much longer follow-up to capture. Our design involved exposing all participants to the same messages from two efficacy domains (response efficacy, self-efficacy), which is suitable for an early-phase, mechanism-focused EMA study; however, people who smoke may require different messages and types of support as they go through the quitting process. Future research should investigate adaptive, stage-matched, or behavior-responsive message systems that tailor message content and delivery based on user data, including real-time user engagement.

This study provides encouraging evidence that delivering daily efficacy messages via text or email may help Mexican-origin Latino/a smokers engage in intermediary behaviors, such as forgoing cigarettes, that can lead to cessation attempts even among those not currently motivated or attempting to quit. The findings underscore the potential value of incorporating principles from Social Impact Theory into intervention designs, particularly by leveraging the influence of former smokers within social networks, as this subgroup may be especially responsive to such interventions. Moreover, because Latino/a populations from low SES groups often encounter structural and cultural barriers to accessing evidence-based cessation services, digital delivery of brief efficacy messages, including those that activate quit support within existing networks, may complement and even overcome barriers that characterize current clinic-based cessation strategies.

Conclusion

This exploratory EMA study provides evidence that social network characteristics may influence smoking cessation-related behaviors among Mexican-origin adults with relatively low education. Having more females and former smokers in one’s social network increases the likelihood of forgoing cigarettes, an important early indicator of cessation. Moreover, individuals in tighter, more supportive networks report more conversations about quitting. Our initial findings suggest that completing more prior EMA surveys is associated with forgoing cigarettes, indicating potential benefits of repeated exposure to simple efficacy messages. Together, these findings suggest that brief digital efficacy messages, combined with supportive social network structures, may promote early steps toward smoking cessation among Latino/a smokers. Overall, this work provides guidance for developing network-based strategies that leverage key social influences to promote smoking cessation among Latinos/a.

Supplemental Material

Supplemental Material - Influence of Social Network Characteristics on Smoking Cessation Conversations and Forgoing Cigarettes: An Ecological Momentary Assessment Study in Mexico and the US

Social Network Determinants of Cigarette Forgoing and Quit-Related Conversations in a 15-Day Ecological Momentary Assessment Study in Mexico and the US by Lizeth Cruz-Jiménez, Kevin A. Carson, James F. Thrasher, Katia Gallegos, Dèsirée Vidaña-Pérez, Diego F. Leal in Tobacco Use Insights.

Footnotes

Acknowledgments

Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health (R01 CA215466-S01 and R01 CA290828). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

ORCID iDs

Lizeth Cruz-Jiménez

Kevin A. Carson

James F. Thrasher

Katia Gallegos-Carrillo

Dèsirée Vidaña-Pérez

Diego F. Leal

Ethical Considerations

All study procedures were approved by the University of South Carolina Institutional Review Board (Protocol No. Pro00102530).

Consent to Participate

Participants provided electronic informed consent before study enrollment.

Author Contributions

Lizeth Cruz-Jiménez contributed to conceptualization, methodology, formal analysis, investigation, writing–original draft, writing–review and editing. Kevin A. Carson contributed to formal analysis, writing–original draft, and writing–review and editing. James F. Thrasher contributed to conceptualization, methodology, writing–review and editing, and supervision. Katia Gallegos contributed to the writing–review, and editing. Dèsirée Vidaña-Pérez contributed to the writing–review and editing. Diego F. Leal contributed to conceptualization, methodology, formal analysis, investigation, writing–original draft, writing–review, editing, and supervision.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Cancer Institute of the National Institutes of Health (R01 CA215466-S01 and R01 CA290828). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online.

Note

References

US Census Bureau . New estimates highlight differences in growth between the U.S. Hispanic and non-Hispanic populations. 2024. https://www.census.gov/newsroom/press-releases/2024/population-estimates-characteristics.html. Accessed 10 October 2024.

Levinson

Pérez-Stable

Espinoza

Flores

Byers

. Latinos report less use of pharmaceutical aids when trying to quit smoking. Am J Prev Med. 2004;26(2):105-111. doi:10.1016/j.amepre.2003.10.012.

Babb

Malarcher

Asman

, et al. Disparities in Cessation Behaviors Between Hispanic and Non-Hispanic White Adult Cigarette Smokers in the United States, 2000–2015. Prev Chronic Dis. 2020;17:190279. doi:10.5888/pcd17.190279.

Lee

Contrera Avila

Ahluwalia

. Differences in cessation attempts and cessation methods by race/ethnicity among US adult smokers, 2016–2018. Addict Behav. 2023;137:107523. doi:10.1016/j.addbeh.2022.107523.

Assari

Sheikhattari

. Social determinants of successful smoking cessation: an eight-year analysis of Population Assessment of Tobacco and Health (PATH) data. J Biomed Life Sci. 2024;4(2):60-70. doi:10.31586/jbls.2024.1070.

Mijs

JJB

Roe

. Is America coming apart? Socioeconomic segregation in neighborhoods, schools, workplaces, and social networks, 1970–2020. Sociol Compass. 2021;15(6):e12884. doi:10.1111/soc4.12884.

Smith

Christakis

. Social Networks and Health. Annu Rev Sociol. 2008;34(1):405-429. doi:10.1146/annurev.soc.34.040507.134601.

Valente

Pitts

. An Appraisal of Social Network Theory and Analysis as Applied to Public Health: Challenges and Opportunities. Annu Rev Public Health. 2017;38(1):103-118. doi:10.1146/annurev-publhealth-031816-044528.

Guilbeault

Centola

. Networked collective intelligence improves dissemination of scientific information regarding smoking risks. PLoS One. 2020;15(2):e0227813. doi:10.1371/journal.pone.0227813. Zollo F, ed.

10.

Christakis

Fowler

. The Collective Dynamics of Smoking in a Large Social Network. N Engl J Med. 2008;358(21):2249-2258. doi:10.1056/NEJMsa0706154.

11.

Blok

de Vlas

van Empelen

van Lenthe

. The role of smoking in social networks on smoking cessation and relapse among adults: A longitudinal study. Prev Med (Baltim). 2017;99:105-110. doi:10.1016/j.ypmed.2017.02.012.

12.

Moore

Teixeira

Stewart

. Effect of Network Social Capital on the Chances of Smoking Relapse: A Two-Year Follow-up Study of Urban-Dwelling Adults. Am J Public Health. 2014;104(12):e72-e76. doi:10.2105/AJPH.2014.302239.

13.

Takagi

Yokouchi

Hashimoto

. Smoking behavior prevalence in one’s personal social network and peer’s popularity: A population-based study of middle-aged adults in Japan. Soc Sci Med. 2020;260:113207. doi:10.1016/j.socscimed.2020.113207.

14.

Gondal

. Diffusion of innovations through social networks: Determinants and implications. Sociol Compass. 2023;17(5):e13084. doi:10.1111/soc4.13084.

15.

Latané

. The psychology of social impact. Am Psychol. 1981;36(4):343-356. doi:10.1037/0003-066X.36.4.343.

16.

Latané

. Dynamic Social Impact: The Creation of Culture by Communication. J Commun. 1996;46(4):13-25. doi:10.1111/j.1460-2466.1996.tb01501.x.

17.

Kitts

Shi

. Toward an Analytic Framework of Social Influence: Behavioral Diffusion. 2018. https://people.umass.edu/jkitts/pubs/NationalAcademyofSciences_2018.pdf. Accessed 21 January 2026.

18.

Krackhardt

Stern

. Informal Networks and Organizational Crises: An Experimental Simulation. Soc Psychol Q. 1988;51(2):123-140. doi:10.2307/2786835.

19.

Bojanowski

Corten

. Measuring segregation in social networks. Soc Networks. 2014;39(1):14-32. doi:10.1016/J.SOCNET.2014.04.001.

20.

Leal

. Locating Cultural Holes Brokers in Diffusion Dynamics Across Bright Symbolic Boundaries. Sociol Methods Res. 2025;00491241251322517. doi:10.1177/00491241251322517. Published online March 19.

21.

Southwell

Yzer

. When (and Why) Interpersonal Talk Matters for Campaigns. Communication Theory. 2009;19(1):1-8. doi:10.1111/j.1468-2885.2008.01329.x.

22.

Southwell

Yzer

. The roles of interpersonal communication in mass media campaigns. Ann Int Commun Assoc. 2007;31:419-462. In: Beck CS, ed. Communication Yearbook. Lawrence Erlbaum Associates.

23.

Morgan

Golden

Noar

, et al. Conversations about pictorial cigarette pack warnings: Theoretical mechanisms of influence. Soc Sci Med. 2018;218:45-51. doi:10.1016/j.socscimed.2018.09.063.

24.

Lambert

Davis

Popova

Thrasher

. Cessation Conversations and Quit Attempts: Differences by Ethnicity and Language Preference. Am J Health Behav. 2020;44(4):473-487. doi:10.5993/AJHB.44.4.9.

25.

Thrasher

Abad-Vivero

Huang

, et al. Interpersonal communication about pictorial health warnings on cigarette packages: Policy-related influences and relationships with smoking cessation attempts. Soc Sci Med. 2016;164:141-149. doi:10.1016/j.socscimed.2015.05.042.

26.

Van Den Putte

Yzer

Southwell

De Bruijn

Willemsen

. Interpersonal communication as an indirect pathway for the effect of antismoking media content on smoking cessation. J Health Commun. 2011;16(5):470-485. doi:10.1080/10810730.2010.546487.

27.

Morgan

Southwell

Noar

Ribisl

Golden

Brewer

. Frequency and Content of Conversations About Pictorial Warnings on Cigarette Packs. Nicotine Tob Res. 2018;20(7):882-887. doi:10.1093/ntr/ntx180.

28.

Hall

Peebles

Bach

Noar

Ribisl

Brewer

. Social Interactions Sparked by Pictorial Warnings on Cigarette Packs. Int J Environ Res Public Health. 2015;12(10):13195-13208. doi:10.3390/IJERPH121013195.

29.

Thrasher

Islam

Davis

, et al. Testing Cessation Messages for Cigarette Package Inserts: Findings from a Best/Worst Discrete Choice Experiment. Int J Environ Res Public Health. 2018;15(2):282. doi:10.3390/ijerph15020282.

30.

Thrasher

Anshari

Lambert-Jessup

, et al. Assessing Smoking Cessation Messages with a Discrete Choice Experiment. Tob Regul Sci. 2018;4(2):73-87. doi:10.18001/TRS.4.2.7.

31.

Loud

Lambert

Porticella

Niederdeppe

Thrasher

. Evaluating Cigarette Pack Insert Messages with Tips to Quit. Tob Regul Sci. 2021;7(3):203-209. doi:10.18001/TRS.7.3.5.

32.

Thrasher

Hackworth

Ferguson

, et al. Differential Responses to Cigarette Package Labeling Alternatives Among Adults Who Smoke: Results From a Randomized Trial. Nicotine Tob Res. 2024;26(12):1646-1655. doi:10.1093/ntr/ntae139.

33.

Lambert

Ferguson

Niederdeppe

, et al. Self-reported attention and responses to cigarette package labels at the end of a two-week randomized trial of cigarette package labeling configurations. Tob Induc Dis. 2024;22:1-11. doi:10.18332/tid/189198.

34.

Baumeister

Vohs

Funder

. Psychology as the Science of Self-Reports and Finger Movements: Whatever Happened to Actual Behavior? Perspect Psychol Sci. 2007;2(4):396-403. doi:10.1111/j.1745-6916.2007.00051.x.

35.

Shiffman

Stone

Hufford

. Ecological Momentary Assessment. Annu Rev Clin Psychol. 2008;4(1):1-32. doi:10.1146/annurev.clinpsy.3.022806.091415.

36.

Ferguson

Shiffman

. Using the methods of ecological momentary assessment in substance dependence research--smoking cessation as a case study. Subst Use Misuse. 2011;46(1):87-95. doi:10.3109/10826084.2011.521399.

37.

Pachucki

Leal

. Is having an educationally diverse social network good for health? Network Science. 2020;8(3):418-444. doi:10.1017/nws.2020.14.

38.

Perry

McConnell

Peng

, et al. Social Networks and Cognitive Function: An Evaluation of Social Bridging and Bonding Mechanisms. Gerontologist. 2022;62(6):865-875. doi:10.1093/geront/gnab112. In: Meeks S, ed.

39.

Burt

. Structural Holes: The Social Structure of Competition. 1995. Harvard University Press. Cambridge. https://www.hup.harvard.edu/books/9780674843714.

40.

Thompson

Fong

Boudreau

, et al. Methods of the ITC Four Country Smoking and Vaping Survey, wave 1 (2016). Addiction. 2019;114((suppl Suppl 1):S6-s14. doi:10.1111/add.14528.

41.

Cho

Thrasher

Swayampakala

, et al. Does Reactance against Cigarette Warning Labels Matter? Warning Label Responses and Downstream Smoking Cessation amongst Adult Smokers in Australia, Canada, Mexico and the United States. PLoS One. 2016;11(7):e0159245. doi:10.1371/journal.pone.0159245. In: Gorlova OY, ed.

42.

Partos

Borland

Thrasher

, et al. The predictive utility of micro indicators of concern about smoking: Findings from the International Tobacco Control Four Country study. Addictive Behaviors. 2014;39(8):1235-1242. doi:10.1016/j.addbeh.2014.04.001.

43.

Eghdami

Ahmadkhaniha

Baradaran

Hirbod-Mobarakeh

. Ecological momentary interventions for smoking cessation: a systematic review and meta-analysis. Soc Psychiatry Psychiatr Epidemiol. 2023;58(10):1431-1445. doi:10.1007/s00127-023-02503-2.

44.

McClure

Heffner

Krakauer

Mun

Klasnja

Catz

. Feasibility, Acceptability, and Potential Impact of a Novel mHealth App for Smokers Ambivalent About Quitting: Randomized Pilot Study. JMIR Mhealth Uhealth. 2023;11:e46155. doi:10.2196/46155.

45.

Businelle

Benson

Hébert

, et al. Project phoenix: Pilot randomized controlled trial of a smartphone-delivered intervention for people who are not ready to quit smoking. Drug Alcohol Depend. 2024;260:111351. doi:10.1016/j.drugalcdep.2024.111351.

46.

Read

Gorman

. Gender and Health Inequality. Annu Rev Sociol. 2010;36(1):371-386. doi:10.1146/annurev.soc.012809.102535.

47.

Foraker

Patten

Lopez

Croghan

Thomas

. Beliefs and attitudes regarding smoking among young adult Latinos: A pilot study. Prev Med (Baltim). 2005;41(1):126-133. doi:10.1016/j.ypmed.2004.10.018.

48.

Khanal

Miani

Finne

Zielke

Boeckmann

. Effectiveness of behavior change interventions for smoking cessation among expectant and new fathers: findings from a systematic review. BMC Public Health. 2023;23(1):1812. doi:10.1186/s12889-023-16713-5.

49.

Faseru

Richter

Scheuermann

Park

. Enhancing partner support to improve smoking cessation. Cochrane Database Syst Rev. 2018;2018(9):CD002928. doi:10.1002/14651858.CD002928.pub4.

50.

Burgess-Hull

Roberts

Piper

Baker

. The social networks of smokers attempting to quit: An empirically derived and validated classification. Psychol Addict Behav. 2018;32(1):64-75. doi:10.1037/adb0000336.

51.

Yuan

Westmaas

Thrul

Toussaert

Hilton

White

. Effectiveness of Peer-Support Interventions for Smoking Cessation: A Systematic Review and Meta-analysis. Nicotine and Tobacco Research. 2023;25(9):1515-1524. doi:10.1093/ntr/ntad059.

52.

Mercken

Snijders

TAB

Steglich

de Vries

. Dynamics of adolescent friendship networks and smoking behavior: Social network analyses in six European countries. Soc Sci Med. 2009;69(10):1506-1514. doi:10.1016/j.socscimed.2009.08.003.

53.

Westmaas

Bontemps-Jones

Bauer

. Social support in smoking cessation: Reconciling theory and evidence. Nicotine Tob Res. 2010;12(7):695-707. doi:10.1093/ntr/ntq077.

54.

Kitts

Leal

. What is(n’t) a friend? Dimensions of the friendship concept among adolescents. Social Networks. 2021;66:161-170. https://linkinghub.elsevier.com/retrieve/pii/S0378873321000125.

55.

Suomi

Liu

. Knowledge sharing in online smoking cessation communities: a social capital perspective. Internet Research. 2022;32(7):111-138. doi:10.1108/INTR-03-2020-0113.

56.

Lakon

Zheng

Pechmann

. Social network tie functions of social support and social influence and adult smoking abstinence. PLoS One. 2024;19(3):e0296458. doi:10.1371/journal.pone.0296458. In: Hensel DJ, ed.

57.

Small

Brant

Fekete

. The Avoidance of Strong Ties. Am Sociol Rev. 2024;89(4):615-649. doi:10.1177/00031224241263602.

58.

Soulakova

Tang

Leonardo

Taliaferro

. Motivational Benefits of Social Support and Behavioural Interventions for Smoking Cessation. J Smok Cessat. 2018;13(4):216-226. doi:10.1017/jsc.2017.26.

59.

Arillo-Santillán

Rodríguez-Bolaños

Gallegos-Carrillo

Monterrubio-Flores

Vidaña-Pérez

Thrasher

. Conversations about cigarette health warning labels and smoking cessation attempts in Mexico. Salud Publica Mex. 2024;67(1):74-82. doi:10.21149/16044.

60.

Lara

Gamboa

Kahramanian

Morales

Hayes

BDE

. Acculturation and Latino health in the United States: a review of the literature and its sociopolitical context. Annu Rev Public Health. 2005;26(1):367-397. doi:10.1146/annurev.publhealth.26.021304.144615.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.85 MB