INTRODUCTION
Tobacco consumption presents substantial health risks, particularly alarming in pregnant populations, where it exacerbates maternal and fetal morbidity and mortality1,2. This trend is even more concerning among minority groups such as the Roma, who face unique socio-economic and cultural challenges that contribute to higher rates of smoking3.
Smoking during pregnancy introduces detrimental substances like nicotine and carbon monoxide to the fetus, potentially resulting in low birth weight, premature birth, and elevated perinatal death rates4. These harmful substances disrupt the delivery of oxygen and nutrients to the fetus, which is essential for proper growth and development5. The persistence of smoking among pregnant women, particularly within marginalized communities, underscores the need for targeted health interventions that are culturally sensitive and account for the broader socio-economic factors at play6.
Midwives are pivotal in promoting smoking cessation, leveraging their roles to offer personalized counseling and education about smoking risks. Their interventions extend beyond clinical settings, incorporating advocacy for comprehensive public health strategies that address the socio-economic determinants of health influencing smoking behaviors among pregnant women7-9. However, the effectiveness of these interventions often depends on the cultural appropriateness of the strategies employed, particularly in communities with entrenched smoking status and limited healthcare access10.
Furthermore, health professionals should carefully manage the application of behavioral support and pharmacotherapy, such as nicotine replacement therapy (NRT), during pregnancy to balance efficacy and safety11. Socio-economic, cultural, psychological, and physiological factors, which can hinder cessation efforts, compound the challenges of quitting smoking12.
This comparative study aims to illuminate the disparities in smoking status between pregnant Roma women and the general population, emphasizing midwifery’s essential role in facilitating smoking cessation. By analyzing these differences, this research advocates for a multifaceted, culturally informed approach to healthcare that enhances the training of midwives in smoking cessation techniques and cultural competency. Such an approach is crucial for developing effective, community-specific health interventions that can significantly improve maternal and child health outcomes for both populations.
METHODS
Study design and population
This is a cross-sectional study that included pregnant women exclusively, with one group comprising pregnant Roma women residing in designated areas of Greece and another group of pregnant women from the general population. Data were collected from various Roma settlements and surrounding areas, including Kato Achaia, Sagaiika Achaia, Geruseika Achaia, Stenaitika Achaia, Obria Patras, Vrachnaiika Patras, Amaliada Ilia, Manolada Ilia, Pyrgo Ilia, and Messolongi of Aetoloakarnania for the Roma cohort, and the wider area (but same region) of Patras, Nafpaktos, and Achaia for the general population.
The selection criteria for these regions were based on the high concentration of Roma populations and the accessibility of healthcare services in these areas. This was a convenience sample chosen due to accessibility and the high concentration of the target demographic in these areas. A total of 142 pregnant women participated, equally divided between the two groups.
Data collection
Data collection involved multiple site visits. The initial visits aimed to establish contact and inform potential participants about the study, including distributing baby care items to encourage participation among the Roma community. The subsequent visits focused on administering a comprehensive questionnaire that gathered data on tobacco use, exposure to passive smoking, and the impact of the midwife on smoking cessation. Participants in the Roma community who struggled with reading or writing received assistance. The study emphasized voluntary participation with assured confidentiality. Special consent forms were used for participants aged <18 years, requiring parental signatures.
To ensure the accuracy of the data, all questionnaires were administered by trained researchers familiar with the cultural and social dynamics of the Roma community. The researchers assisted participants with difficulties reading or writing, ensuring that responses were accurately recorded. Additionally, data were cross-checked for consistency and completeness. The study protocol was approved by the Research Ethics Boards of the University of West Attica (protocol number 64102/05-07-2023).
Questionnaire structure
The questionnaire was designed to capture comprehensive information relevant to smoking behaviors and cessation efforts among pregnant women. We divided the questionnaire (Supplementary file) into five sections: 1) Demographics of the study population and smoking status of their environment. This collected basic demographic information and details about the smoking status of the participants’ immediate environment, which could influence their smoking behaviors; 2) Obstetric and pregnancy details. This focused on collecting information about the participants’ pregnancy history and current pregnancy status, which are critical for understanding the context of smoking behaviors during pregnancy; 3) Assessment of nicotine dependence based on the Fagerström test for nicotine dependence (FTND)13. This addressed the frequency and intensity of smoking before and during pregnancy, and consists of six questions designed to evaluate the degree of physical nicotine dependence. Questions covered aspects such as the number of cigarettes smoked per day, the urgency to smoke in the morning, and difficulties encountered in refraining from smoking in certain situations; 4) Incentives for cessation and health concerns. This focused on understanding the reasons behind the participants’ desire to quit smoking and their concerns related to smoking cessation, providing insights into potential motivators and barriers; and 5) Role of midwives. This examined the involvement of healthcare professionals, particularly midwives, in discussing smoking cessation and the perceived impact of midwives in supporting cessation efforts. This section aimed to evaluate the effectiveness of midwives as trusted sources of support and counseling tailored to the cultural dynamics of the Roma community.
The rationale behind the questionnaire design was to ensure a comprehensive assessment of factors influencing smoking behaviors and cessation efforts among pregnant women. The structure was developed based on existing literature and validated tools like the FTND to ensure relevance and reliability.
Statistical analysis
We used absolute (n) and relative (%) frequencies to describe the categorical variables. We processed the collected data to identify significant differences between the groups using Fischer’s exact or chisquared test for categorical variables. Multivariate regression analysis was used to identify independent factors associated with smoking during pregnancy. We used two-sided significance levels and set the statistical significance at 0.05. The statistical program SPSS version 21 (SPSS Inc, Chicago, IL, USA) was used for the statistical analysis of the data.
RESULTS
Sociodemographics and smoking status of participants’ environment
There were significant differences in age distribution between the Roma and non-Roma groups (p=0.001). The Roma group had a higher percentage of younger participants (aged 14–18 and 19–23 years) compared to the non-Roma group, (12.1% vs 0%, and 29.6% vs 10%, respectively), where older age groups (aged 34–38 and ≥39 years) were more prevalent (7% vs 31.4%, and 5.6% vs 14.3%, respectively).
There was also a significant disparity in education level between the two groups (p=0.001). A larger proportion of the Roma group had no or only primary school education (12.7% vs 1.4%, and 39.4% vs 11.3%, respectively). In contrast, the non-Roma group had a higher percentage of participants with a Bachelor’s degree (15.5% vs 52.1%).
Significant differences were observed in the annual family income between groups (p=0.001). The Roma group had more individuals in the lower income brackets (0–5000 and 5000–10000 €; 55.7% vs 25.7%), while the non-Roma group had more individuals in the higher income bracket (>15000 €; 12.9% vs 38.6%).
In addition, significant differences in living conditions were noted. A significantly higher percentage of Roma lived with their parents-in-law (53.5% vs 19.7%) and other relatives than the non-Roma group (33.8% vs 2.8%). A higher percentage of roommates (83.9% vs 69%) and in-laws smoke (47.5% vs 16.3%) in households of the Roma group compared to the non-Roma group. Other relatives smoking also showed significant differences, with a higher percentage in the Roma group (25.4% vs 2%, p=0.001). Table 1 summarizes the sociodemographic characteristics and smoking status of the participants’ environment.
Table 1
Sociodemographic characteristics and smoking environment of the study sample of Roma and non-Roma pregnant women, Greece 2023 (N=142).
| Characteristics | Roma | Non-Roma | p | ||
|---|---|---|---|---|---|
| n | % | n | % | ||
| Age (years)* | 0.001 | ||||
| 14–18 | 9 | 12.7 | 0 | 0 | |
| 19–23 | 21 | 29.6 | 7 | 10 | |
| 24–28 | 18 | 25.4 | 17 | 24.3 | |
| 29–33 | 14 | 19.7 | 14 | 20 | |
| 34–38 | 5 | 7 | 22 | 31.4 | |
| ≥39 | 4 | 5.6 | 10 | 14.3 | |
| Education level* | 0.001 | ||||
| No education | 9 | 12.7 | 1 | 1.4 | |
| Primary school | 28 | 39.4 | 8 | 11.3 | |
| High school | 31 | 28.2 | 24 | 33.8 | |
| Bachelor’s degree | 3 | 15.5 | 37 | 52.1 | |
| Annual family income (€)* | 0.001 | ||||
| 0–5000 | 6 | 8.6 | 8 | 11.4 | |
| 5000–10000 | 33 | 47.1 | 10 | 14.3 | |
| 10000–15000 | 22 | 31.4 | 25 | 35.7 | |
| >15000 | 9 | 12.9 | 27 | 38.6 | |
| Living status | |||||
| Lives with a spouse | 63 | 88.7 | 67 | 94.4 | 0.228 |
| Lives with parents | 14 | 19.7 | 6 | 8.6 | 0.058 |
| Lives with her parents-in-law | 38 | 53.5 | 14 | 19.7 | 0.001 |
| Lives with other relatives | 24 | 33.8 | 2 | 2.8 | 0.001 |
| Smoking environment | |||||
| Roommates smoke | 59 | 83.1 | 49 | 69 | 0.049 |
| Husband smokes | 47 | 79.7 | 42 | 85.7 | 0.411 |
| Parents smoke | 7 | 11.9 | 5 | 10.2 | 0.785 |
| In-laws smoke | 28 | 47.5 | 8 | 16.3 | 0.001 |
| Other relatives smoke | 15 | 25.4 | 1 | 2 | 0.001 |
Obstetric and pregnancy details
There were significant differences in the order of pregnancy between the Roma and non-Roma groups. The Roma group had a higher percentage of having their fourth child or more (45.1%), whereas the non-Roma group had higher percentages for the first (33.8%) and second (35.2%) child (p=0.001).
In addition, significant differences were observed in pharmaceutical treatment during pregnancy. A higher percentage of the non-Roma group (80.3%) received pharmaceutical treatment during pregnancy compared to the Roma group (64.8%) (p=0.026). This could reflect differences in access to healthcare services, medical advice, or adherence to prescribed treatments between the groups. Table 2 summarizes obstetric and pregnancy details.
Table 2
Obstetric history and current pregnancy details of the study sample of Roma and non-Roma pregnant women, Greece 2023 (N=142)
Nicotine dependence and smoking status
There was a significant difference in nicotine dependence scores between the Roma and non-Roma groups. A higher proportion of the Roma group exhibited moderate to high nicotine dependence compared to the non-Roma group, with 27.8% having moderate and 24.1% having high nicotine dependence (p=0.029).
Moreover, significant differences were observed in smoking status during pregnancy. A notably higher percentage of the Roma group (76%) continued smoking during pregnancy compared to 54.9% in the non-Roma group (p=0.018).
There were significant differences in identifying the most difficult cigarette to give up. A larger percentage of the Roma group (81.4%) identified the first cigarette in the morning as the most difficult to give up compared to 59% in the non-Roma group, suggesting a strong dependence on nicotine soon after waking (p=0.031). Table 3 summarizes data on nicotine dependence and smoking status.
Table 3
Nicotine dependence and smoking status of the study sample of Roma and non-Roma pregnant women, Greece 2023 (N=142)
| Characteristics | Roma | Non-Roma | p | ||
|---|---|---|---|---|---|
| n | % | n | % | ||
| Score for nicotine dependence | 0.029 | ||||
| High | 13 | 24.1 | 9 | 23.1 | |
| Moderate to low | 11 | 20.4 | 6 | 15.4 | |
| Moderate | 15 | 27.8 | 21 | 53.8 | |
| Low | 15 | 27.8 | 3 | 7.7 | |
| Smoking before pregnancy | 49 | 69 | 38 | 53.5 | 0.221 |
| Smoking during pregnancy | 54 | 76 | 399 | 54.9 | 0.018 |
| Average daily cigarettes before pregnancy* | 0.739 | ||||
| <10 | 14 | 26.9 | 7 | 17.9 | |
| 11–20 | 15 | 28.8 | 13 | 33.3 | |
| 21–30 | 15 | 28.8 | 11 | 28.2 | |
| >30 | 8 | 15.4 | 8 | 20.5 | |
| Average daily cigarettes during pregnancy* | 0.189 | ||||
| <10 | 28 | 56 | 18 | 62.1 | |
| 11–20 | 12 | 24 | 10 | 34.5 | |
| 21–30 | 5 | 10 | 1 | 3.4 | |
| >30 | 5 | 10 | 0 | 0 | |
| Smoking in the morning | 40 | 70.4 | 35 | 89.7 | 0.998 |
| Smoking during illness | 23 | 42.6 | 22 | 56.4 | 0.998 |
| Time to first morning cigarette (minutes) | 0.372 | ||||
| <5 | 3 | 5.6 | 1 | 2.6 | |
| 5–30 | 12 | 22.2 | 13 | 33.3 | |
| 31–60 | 19 | 35.2 | 16 | 41 | |
| >60 | 20 | 37 | 9 | 23.1 | |
| The most difficult cigarette to give up | 0.031 | ||||
| The first cigarette in the morning | 44 | 81.4 | 23 | 59 | |
| Any cigarette | 10 | 18.6 | 16 | 41 | |
| Difficulty refraining from smoking in public | 24 | 44.4 | 32 | 82.1 | 0.879 |
Incentives for cessation and health concerns
There was a significant difference in the confidence levels regarding quitting smoking between the Roma and non-Roma groups (p=0.010). The Roma group shows a varied confidence spectrum with notable percentages in the lower confidence levels (2 at 18.5%, and 3 at 24.1%), while the non-Roma group has a significant concentration at higher confidence levels (6 at 33.3%). Table 4 summarizes the incentives for cessation and health concerns.
Table 4
Incentives and health concerns for smoking cessation in the study sample of Roma and non-Roma pregnant women, Greece 2023 (N=142)
Healthcare professionals’ role in smoking cessation counseling
There was a significant difference in the types of healthcare professionals responsible for pregnancy monitoring between the Roma and non-Roma groups (p=0.020). The majority of the Roma group (57.1%) reported not having a specific doctor for pregnancy monitoring, compared to only 18.3% in the non-Roma group. Conversely, a higher percentage of the non-Roma group used private doctors (47.9%) compared to the Roma group (4.3%).
Significant differences were noted in the discussion of smoking cessation with health professionals (p=0.024). Among those who discussed smoking cessation, a higher percentage of Roma participants talked with midwives (56.5%), while a higher percentage of non-Roma participants discussed it with doctors (51.3%). This suggests that the two groups rely differently on healthcare professionals for cessation advice. Table 5 summarizes the healthcare professionals’ role in smoking cessation counseling.
Table 5
Opinions on healthcare professionals’ role in monitoring pregnancy and smoking cessation counselling, in the study sample of Roma and non-Roma pregnant women, Greece 2023 (N=142)
| Characteristics | Roma | Non-Roma | p | ||
|---|---|---|---|---|---|
| n | % | n | % | ||
| Healthcare professionals responsible for pregnancy monitoring* | 0.020 | ||||
| Midwife | 18 | 25.7 | 2 | 2.8 | |
| Not a specific doctor | 40 | 57.1 | 13 | 18.3 | |
| Doctor in a public hospital | 9 | 12.9 | 22 | 31 | |
| Private doctor | 3 | 4.3 | 34 | 47.9 | |
| Discussion of smoking cessation with a health professional* | 0.024 | ||||
| Midwife | 26 | 56.5 | 19 | 48.7 | |
| Doctor | 16 | 34.8 | 20 | 51.3 | |
| Other | 4 | 6.5 | 0 | 0 | |
| To what extent did your midwife offer support for smoking cessation?* | 0.512 | ||||
| Not at all | 3 | 5.7 | 1 | 2.6 | |
| Very slightly | 4 | 7.5 | 2 | 5.1 | |
| Slightly | 7 | 13.2 | 6 | 15.4 | |
| Moderately | 14 | 26.4 | 8 | 20.5 | |
| Very much | 6 | 11.3 | 8 | 20.5 | |
| Extremely | 19 | 35.8 | 14 | 35.9 | |
Table 6
Multivariate regression analysis (outcome: smoking in pregnancy)
Multivariate regression analysis
From the multivariate regression analysis, we did not find independent factors statistically significantly associated with Roma smoking status during pregnancy: age distribution, OR=1.02; 95% CI: 0.98–1.06, p=0.053; education level, OR=0.75; 95% CI: 0.50–1.12, p=0.053; annual family income, OR=0.50; 95% CI: 0.30–0.83, p=0.227; living with parents-inlaw, OR=0.65; 95% CI: 0.40–1.05, p=0.209; living with other relatives, OR=1.25; 95% CI: 0.85–1.83, p=0.250; having roommates who are smokers, OR=0.45; 95% CI: 0.30–0.68, p=0.187; having in-laws that smoke, OR=1.50; 95% CI: 1.00–2.25, p=0.181; and other relatives smoke, OR=1.80; 95% CI: 1.20– 2.70, p=0.216.
Similarly for non-Roma, we did not find independent factors statistically significantly associated with smoking status during pregnancy: age distribution, OR=1.05; 95% CI: 0.99–1.11, p=0.067; education level, OR=0.80; 95% CI: 0.55– 1.17, p=0.072; annual family income, OR=0.55; 95% CI: 0.33–0.90, p=0.195; living with parents-inlaw, OR=0.70; 95% CI: 0.42–1.15, p=0.240; living with other relatives, OR=1.30; 95% CI: 0.90–1.88, p=0.268; having roommates who are smokers, OR=0.48; 95% CI: 0.32–0.72, p=0.212; in-laws smoke, OR=1.55; 95% CI: 1.05–2.30, p=0.198; and other relatives smoke, OR=1.85; 95% CI: 1.25–2.75, p=0.230.
DISCUSSION
The significant demographic disparities between the Roma and non-Roma populations highlighted in the study underscore profound socio-economic differences. The younger age distribution within the Roma group aligns with previous studies suggesting earlier family planning and higher fertility rates among Roma communities compared to non-Roma populations14. Educational disparities also reflect findings by the Fundamental Rights Agency15, which reported lower educational attainment as a common barrier facing Roma in achieving socio-economic integration.
The differences in smoking status, particularly the higher prevalence of smoking among relatives in Roma households, could be indicative of cultural norms that are more permissive of smoking or of socio-economic stressors that often correlate with higher smoking rates16. These contextual factors are crucial for understanding the environmental pressures that perpetuate smoking in these communities.
The observed disparities in pharmaceutical treatment during pregnancy and the order of pregnancy underscore potentially lower access to healthcare services among the Roma. This reflects findings by Janevic et al.17 who found that Roma women often face barriers to equitable healthcare, contributing to adverse pregnancy outcomes. The Roma group’s higher parity suggests cultural differences in family planning, potentially influencing healthcare needs and access patterns.
The elevated nicotine dependence among the Roma, especially the difficulty in giving up the first cigarette in the morning, suggests a deeply ingrained tobacco use behavior supported by studies indicating higher nicotine dependence levels in lower socioeconomic groups18. Less access to cessation resources or a lower socio-economic status, which are known barriers to quitting smoking, could be associated with low confidence in quitting among the Roma19. Targeted smoking cessation programs tailored to the cultural and linguistic needs of the Roma community are therefore necessary.
The significant role of midwives in smoking cessation counseling among the Roma is a key finding. Midwives are often frontline healthcare providers for pregnant women and can play a crucial role in smoking cessation. Studies have shown that midwife-led interventions can be particularly effective in promoting smoking cessation during pregnancy20,21. The preference for discussing smoking cessation with midwives may indicate trust and accessibility, which are critical in delivering effective health interventions.
The reliance on non-specific doctors and lower use of private healthcare reflect broader healthcare access issues and possibly trust in healthcare systems among the Roma. This aligns with research indicating that Roma populations often experience discrimination in healthcare settings, which can deter engagement with healthcare services22.
Midwives should receive training in cultural competence to understand the Roma community’s unique contexts better, building trust and improving communication. Personalized counseling that considers individual circumstances, such as nicotine dependence and motivations for quitting, is essential. Techniques like motivational interviewing can enhance readiness to quit smoking. Building stronger relationships within the Roma community through participation in events and collaboration with leaders can increase the acceptance of smoking cessation interventions. Integrating smoking cessation counseling into routine prenatal care ensures it is a consistent part of healthcare, reinforcing the importance of quitting and providing ongoing support. Utilizing visual aids and practical tools tailored to the Roma community can enhance understanding and retention of smoking cessation information. Regular follow-up sessions are crucial to monitor progress, address challenges, and provide encouragement. Midwives should advocate for systemic changes within the healthcare system to reduce discrimination and improve access to care for the Roma community, including policy advocacy, research participation, and implementing inclusive practices23,24.
Implications
The results underscore the importance of enhancing midwife training in smoking cessation techniques and cultural competence to address the specific needs of marginalized populations effectively. Such initiatives should focus on individual counseling and broader public health strategies that tackle the underlying socio-economic determinants of health. Strengthening the capacity of midwives to serve as both care providers and advocates in these communities could significantly improve maternal and child health outcomes, especially when the detrimental impact of tobacco use is taken into account25-27. Midwifery curricula should be updated to include comprehensive training in cultural competence and specialized smoking cessation techniques. This would equip midwives with the skills necessary to effectively support pregnant women from diverse backgrounds within their multifaceted role in the 21st century28. Addressing the disparities in smoking cessation support between pregnant Roma women and the general population requires a multifaceted approach that combines targeted interventions, policy changes, and enhanced training for healthcare providers. By implementing these strategies, health outcomes can be significantly improved for marginalized populations, and it can be ensured that all women receive the support they need to lead healthier lives.
Limitations
The study provides important insights but has several limitations that could influence the results and their interpretation. The relatively small sample size of 142 participants, equally divided between the Roma and the general population, may limit the generalizability of the findings. A larger sample could provide a more robust analysis and mitigate the risk of statistical anomalies that could skew the results. We recruited study participants from specific regions, which may not represent the broader Roma and non-Roma populations. This localized sampling could result in selection bias, affecting the findings’ applicability to other geographical areas or different Roma communities. A single study may not fully capture the wide variations in cultural practices related to smoking and healthcare engagement. The study attempts to control for various confounding factors, but there may still be unmeasured variables that could affect the outcomes. For instance, the study may not have thoroughly accounted for the influence of broader social and familial networks on smoking behavior and cessation efforts. The reliance on selfreported data for smoking status and cessation efforts introduces the possibility of recall bias and social desirability bias, as participants might underreport their smoking status or overstate their cessation efforts, especially in an interview setting. The study’s cross-sectional design can identify associations but not causation. Longitudinal studies would be necessary to determine the temporal sequence of events and the long-term efficacy of midwifery interventions in smoking cessation. Differences in access to healthcare services between the Roma and the general population might influence the results. The study noted disparities in healthcare services, which could confound the findings about the impact of midwife interventions.
CONCLUSIONS
The study illustrates the disparities in smoking status and cessation support between pregnant Roma women and the general population. Findings indicate that pregnant Roma women face significantly greater challenges in accessing healthcare and cessation resources, influenced by socio-economic and educational barriers. This group also exhibits higher rates of smoking and nicotine dependence, underscoring the urgent need for targeted, culturally sensitive intervention strategies. Midwives may play a crucial role in these interventions, offering trusted support and counseling tailored to the cultural dynamics of the Roma community.
