An evaluation of early marriage and the mental state of Roma women: A cross-sectional study

Abstract

Early marriage (i.e., at less than 18 years of age) is a significant global problem threatening the well-being of women. This cross-sectional study evaluated early marriage and the mental health of Roma women in Adana, Turkey. Data were collected between March 2019 and September 2019 using the Descriptive Characteristics Form and the Brief Symptom Inventory. We interviewed 272 married Roma women over 18 years of age, of whom 59.6% (n = 162) had married before the age of 18 years; of these, 14.2% had done so before the age of 15. Significant positive correlations were found between the duration of the marriage and measures of interpersonal sensitivity, somatization, depression, anxiety, phobia, paranoid ideation, psychoticism, and the Global Severity Index in the women who had married early. Among women who married after 18, the duration of marriage had a significant positive relationship with somatization and phobic anxiety. Early marriage and a longer duration of marriage were associated with more symptoms of psychological distress.

Keywords

early marriage mental health psychological distress Roma women

Introduction

Early marriage is defined as an official or traditional marriage in which one of the partners is younger than 18 years of age (United Nations Children's Fund [UNICEF], 2021). The legal age of marriage is 16–18 years in many countries. However, it has been suggested by UNICEF that the legal age of marriage should be at least 18 (UNICEF, 2022). Traditional marriages without official sanction before the age of 18 are common worldwide (Arthur et al., 2018; World Health Organization [WHO], 2016). Globally, 21% of women are married before the age of 18. In more than one-third of these marriages, the age of marriage of the woman is less than 15 years (UNICEF, 2012, 2014a, 2014b). For men worldwide, 4.5% marry when they are below the age of 18 (Gaston et al., 2019; United Nations Population Fund [UNFPA], 2012). Although early marriages involve both genders, young women are at a greater risk than young men of sexual, physical, and psychological violence and related outcomes (UNICEF, 2012; WHO, 2016). In recent years, policymakers around the word have made efforts to reduce the frequency of early marriages (Council of Europe, 2015a; UNICEF, 2021). However, despite such attempts, early marriage remains a significant public health concern in many countries. Early marriage is particularly prevalent in some cultures and countries, including those in central Europe, eastern Europe, Asia, and Africa (Gaston et al., 2019; UNICEF, 2012; WHO, 2016). Roma people, who have a large population in Romania, Italy, Ireland, Poland, Bulgaria, Greece, Serbia, Moldova, and Turkey, are distinct cultural minorities in which early marriages frequently occur (Council of Europe, 2015a). The prevalence of early marriage in the Roma population is about twice that of the general population in these countries (Council of Europe, 2015a, 2015b; Sedlecky & Rašević, 2015).

In Turkey, rates of early marriages among Roma girls are well above the Turkish average. Almost half of Roma girls under 18 are married, and 13.6% of those younger than 15 are married (Tanrıverdi et al., 2012; Taylan, 2016; Turkey Ministry of Family and Social Policies General Directorate of Family and Community Services [TMFSPG], 2020). The average age of marriage for a Roma girl in Turkey is between 13 and 17 years (Özkan, 2006). It is further estimated that between 25%–30% of Roma women aged 15–19 years in Romania are married (Council of Europe, 2015a, 2015b).

Cultural and socioeconomic factors may explain the phenomenon of early marriage in Roma communities. Community pressure and traditional values influence parents to accept early marriage for their children (Sedlecky & Rašević, 2015). Values and rituals, as well as power relations, are clearly defined in the Roma culture (Casey, 2014; Sedlecky & Rašević, 2015). In many Roma families, grandparents and parents are the highest authority, and their decisions are rarely opposed. Parents’ views define the boundary between childhood and adulthood. The acceptance of a teenage girl into the Roma social environment is carried out through the assimilation of sexual roles and the culture of marital and family relationships. Marriage, and later motherhood, confer status on Roma women. The emphasis on building a family early, and other traditions related to sexuality, leads many Roma women to consider early marriage to be a normal part of their life course. This practice ensures they have respect and relevance within their communities and allows Roma girls to consider early marriage as normative. In general, Roma men also tend to marry younger than the general population, and therefore the age difference between spouses is usually small. The notion of the normality of early marriage in Roma communities is somewhat controversial. While some Roma boys and girls support the practice, others may be critical of it (Afanasieva et al., 2020; Stojanovski et al., 2017). Roma parents, girls and boys are all navigating the spaces between the demands of modern society and the requirements of traditions and values. Community pressure and traditional values lead parents to accept early marriage for their children (Afanasieva et al., 2020; Bosnjak & Acton, 2013; Sedlecky & Rašević, 2015; Stojanovski et al., 2017).

The situation for Roma women in Adana, Turkey, is quite similar to that of Roma women in other parts of the world. The Roma people in Turkey are poor in terms of social, economic, political, and civil welfare, and lack decision-making power to influence matters concerning them. They are also more likely to lack access to the opportunities afforded by high income, education, employment, and safe and stable housing (Adana Valiliği, 2019; Aktepe & Atay, 2017; Balkız & Göktepe, 2014; Casey, 2014; Çelik & Yüce Tar, 2015; Danka, 2008; Sedlecky & Rašević, 2015; Tanrıverdi et al., 2012; Voicu & Popescu, 2009). In conditions of exclusion, poverty and marginality, the Roma values and traditions about family generate a situation in which marriage may be the only viable strategy for survival from a very early age. The social structure of Roma society is based on strict gender roles in which Roma women are expected to marry, have children, and care for their husbands and children. Structural inequities for Roma can be further worsened by gender-related inequalities in this population (Casey, 2014; Sedlecky & Rašević, 2015; Tanrıverdi et al., 2012; Voicu & Popescu, 2009). Early marriages cause or exacerbate many difficulties for Roma women. Given these circumstances, Roma women are one of the groups most adversely affected by early marriage, compared to other communities in Turkey and worldwide (Council of Europe, 2015a; Gamella, 2018; Martin & Gamella, 2005; Tanrıverdi et al., 2012). Gender-based violence and early marriage faced by Roma women are viewed by some Roma people as normal cultural characteristics of the Roma community and not as problems in themselves. However, these issues are pervasive and threaten the status and well-being of young women. The tradition of early marriage in particular makes Roma women vulnerable to morbidity and mortality risk factors. A Roma girl is expected to fulfill her gendered role as a mother and a wife (Afanasieva et al., 2020; Bosnjak & Acton, 2013). Adolescent Roma girls who have not yet reached physical and psychosocial maturity are expected to marry, establish a family, and have children. As a result, they end their formal education, become mothers, assume full responsibility for the management of their homes, and start working out of the home in addition to having children earlier than is optimal for their development (Aktepe & Atay, 2017; Council of Europe, 2015a; Gamella, 2018; Tanrıverdi et al., 2012). Increased risk for partner violence and its consequences for women's health are other aspects of early marriage faced by these young women (Hotchkiss et al., 2016; McFarlane et al., 2016; Nasrullah et al., 2014; WHO, 2013).

Roma women are at a higher risk of mental illness than non-Roma women (Foldes & Covaci, 2012). A study conducted in Serbia found that Roma women were more unhappy and dissatisfied with their lives than non-Roma women (Cvjetkovic et al., 2017). According to another study conducted in Romania and Bulgaria, Roma women have more mental health problems than non-Roma (Le Cook et al., 2019). A similar study in Portugal revealed that Roma women experience anxiety, depression, and suicidal ideation, and attempt suicide many times more often than the population average (Pereira et al., 2016).

These mental health disparities may be related to the lack of opportunities to set and achieve goals in their lives and the sense of not belonging to society and family, which are strongly associated with symptoms such as low self-esteem and depression during the transitional stages from childhood to adolescence and from adolescence to young adulthood. In addition, the multiple risk factors such as physical and sexual violence, poverty, and adolescent pregnancy that these women are exposed to from adolescence may worsen over time and lead to psychological stress and depression (Le Cook et al., 2019). In addition to all these social and economic factors, early marriages and their consequences harm the psychosocial health of Roma women (Hajduchová & Urban, 2014; Medas, 2020; Robinson et al., 2022; Vorvolakos et al., 2012).

Improving the mental health of Roma women requires a systematic approach, policy reform, research studies, and evidence-based interventions. Studies should therefore be performed to determine the causes, prevalence, and consequences of mental health issues and the support needs of Roma women who marry before the age of 18 (Le Cook et al., 2019; Svanemyr et al., 2015).

Many studies have been conducted to evaluate the physical and mental health of Roma women and find that they have elevated rates of physical health problems, including obesity, hypertension, diabetes, and coronary artery disease, as well as mental health problems such as major depression, anxiety disorder, and substance abuse (Carrasco-Garrido et al., 2010; Coea & Cvorovic, 2017; Cook et al., 2013; Kotrotsiou et al., 2014; McGorrian et al., 2013; Van Cleemput, 2018). However, few studies have focused on the long-term effects of early marriage on women's mental health, and none to date have addressed this issue among Roma women in Turkey (Le Strat et al., 2011; McFarlane et al., 2016).

The United Nations Convention on the Rights of the Child recommends the protection of the civil, political, economic, social, and cultural rights of every child, regardless of race, religion, or ability (UNICEF, 2021). It further states that every child has the right to survival, protection, and education. Therefore, it is essential to reveal how early marriages affect the mental state of Roma women to plan and implement interventions that can prevent the harmful effects of this practice on public health and economy, and to detect, treat, and encourage women in early marriages to receive appropriate health care. This understanding can also contribute significantly to preventing early marriages (Le Strat et al., 2011; McFarlane et al., 2016; Shives, 2008). Accordingly, this study was conducted to evaluate the relationship between early marriage and the mental state of Roma women in Adana, Turkey.

Methods

Study design and setting

This study used a cross-sectional population survey to examine the prevalence of early marriage and its correlates with mental state among Roma women in Adana, Turkey. The study was carried out in a part of Yüreğir, a small town in Adana, between March and September 2019. Adana is the fifth most populous city in Turkey and is considered the junction of migration in the country. Yüreğir has a high proportion of people with low socioeconomic status and is notable for high levels of migration into the region. Many different ethnic groups, including Roma people, live in Yüreğir, although their interactions are limited.

Participants

About 900 Roma people, including adults and older people, live in Yüreğir (Adana Valiliği, 2019), and the study population was estimated as .5 × 900 = 450 Roma women. The sample size of the study was calculated on the basis of the known universe approach. To represent this population, a power calculation suggested minimum sample size was 269 based on a 95% confidence interval (α = .05), a standard error of 5% (d = .05) and incidence of Roma women of p = .5 and q = .5 (Cochran, 1977). A list of married Roma women was obtained by interviewing the headman of the Roma neighborhood and the head of the Adana Roma Association. Roma women who were married, aged over 18 years old, and had sufficient communication abilities were included in the study. Simple random sampling was performed using a random numbers table, and 272 Roma women were contacted. The participants spoke Turkish well. No data exclusion occurred in our analysis.

Data collection

The study data were collected via face-to-face interviews by the author of the study. Accompanied by the head of the Adana Roma Association, the author met with the participants and shared information about the study process. The survey questions were read aloud as the participants were either illiterate or had low levels of education. The interviews were held in a separate room in the women's homes. Other family members waited in another room. Each interview lasted 15–20 min. The participants were not given any incentives.

The data were gathered using the Descriptive Characteristics Form and the Brief Symptom Inventory (BSI). The Descriptive Characteristics Form was created by the researcher based on the relevant literature. The form was composed of 13 questions about the women's sociodemographic characteristics (age, marital status, marriage age, mother's marriage age, duration of the marriage, employment, social security status, educational status, having children, number of children) and health (chronic disease, name of chronic disease, smoking) (Coea & Cvorovic, 2017; Kotrotsiou et al., 2014; Van Cleemput, 2018).

The BSI is the short form of the Symptom Checklist-90-Revised (SCL-90-R). It was developed by Derogatis (1992) and adapted to Turkish culture by Şahin and Durak (1994). The BSI is a 53-item, 5-point self-report Likert scale that is used to screen various psychological symptoms. Higher scores on the scale indicate a higher frequency of symptoms. Derogatis (1992) reported that Cronbach's alpha for the BSI's nine subscales ranged from .71 to .85. Şahin and Durak (1994) reported that the Cronbach's alpha of the subscales of the Turkish version of the scale ranged from .63 to .86, while the Cronbach's alpha calculated for the total score of the scale was .93–.96.

The BSI consists of nine symptom dimensions, namely, somatization, obsession-compulsion, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychoticism, and three global indices of distress: The Global Severity Index, the Positive Symptom Distress Index, and the Positive Symptom Total. These global indices measure the current or past levels of symptomatology, symptom intensity, and the number of reported symptoms, respectively (Derogatis, 1992; Şahin & Durak, 1994).

Statistical analysis

The data were analyzed using SPSS 21. The Shapiro–Wilk test was used to check the normality of the continuous variables. The data relating to the age of the participants, their mother's age at marriage, the duration of their marriage, and the BSI were normally distributed. Student's t-test was used to compare the women who married before the age of 18 and those who married after the age of 18 and the mean scores of the BSI between the women with chronic diseases and those without chronic diseases. The data on the number of children did not have a normal distribution, so the Mann–Whitney U test was employed to compare the data. A chi-square test was used to compare the categorical variables of the women who married before 18 and those who married after the age of 18. Fisher's exact test was used when the proportion of cells with expected values less than five was greater than 20%. The Pearson correlation coefficient was calculated to examine the linear relationship between the scores for the BSI, age at marriage, and duration of the marriage. Finally, multiple linear regression analysis with backward elimination was applied to analyze the effects of chronic diseases, age at marriage, and duration of marriage on the results of the BSI. The reliability of the BSI was expressed as Cronbach's alpha. The statistical significance was set at p < .05.

Ethical considerations

The study was conducted in accordance with the principles of the Declaration of Helsinki. Ethical approval was obtained from the Cukurova University Ethical Board of Clinical Research (number 79/51, dated July 6, 2018), and written permission was obtained from the city governor. Participation in the study was voluntary, and oral and written informed consent was obtained from the women after being given relevant information about the study, including aspects relating to confidentiality, privacy, and the right to withdraw from the study.

Results

Sociodemographic characteristics of the study participants

The study was conducted with 272 Roma women, of whom 59.6% (n = 162) had married before the age of 18 years, and 40.4% (n = 110) had married after the age of 18 years. The mean age of the women was 37.64 ± 13.62 years and ranged from 18 to 67 years. Of the Roma women in early marriages, 14.2% (n = 22) had married before the age of 15 years, and 85.8% (n = 140) had married when they were between 15 and 17 years of age.

Table 1 shows the sociodemographic characteristics and health knowledge of the participants. The majority of women in the study had a perception of low socioeconomic level; 37.9% of the participants did not receive any education, 40.7% of the women, who have not completed secondary school education, dropped out due to financial reasons. There was no significant difference in terms of education, employment, and socioeconomic status between women who married before and after the age of 18 (p > .05) (Table 1). However, significant differences were found between the women who married before and after the age of 18 in terms of their mean age, mother's age at marriage, and duration of marriage (p < .05) (see Table 1 for results).

Table 1.

Descriptive characteristics of participants according to age at marriage.

	Age at marriage <18 years(n = 162)		Age at marriage ≥18 years(n = 110)		Total(N = 272)
	Mean ± SD	Min–Max	Mean ± SD	Min-Max	Mean ± SD	Min– Max	p
Age	39.98 ± 15.11	18–67	34.21 ± 10.21	18–65	37.64 ± 13.62	18–67	.001^a
Mothers’ age at marriage	16.27 ± 2.16	12–21	16.89 ± 2.00	13–26	16.52 ± 2.12	12–26	.016^a
Duration of marriage	24.19 ± 15.36	1–52	14.70 ± 10.59	0–47	20.35 ± 14.39	0–52	<.001^a
	Median [25%–75%]		Median [25%–75%]		Median [25%–75%]
Number of children	3 [2–4]		2 [2–3]		3 [2–4]		.001^b

	n	%	n	%	N	%
Age (years)
18–25	40	24.7	29	26.4	69	25.4	<.001^c
26–35	35	21.6	39	35.5*	74	27.2
36–45	26	16.0	29	26.4*	55	20.2
46–55	27	16.7*	9	8.2	36	13.2
56–65	34	21.0*	4	3.6	38	14.0
Age at marriage (years)
12–14	22	14.2	0	0	22	8.1
15–17	140	85.8	0	0	140	51.5
18 and above	0	0	110	100.0	110	40.4
Education
Illiterate	69	42.6	34	30.9	103	37.9	.069^c
Literate	38	23.5	30	27.3	68	25.0
Primary school	42	25.9	41	37.3	83	30.5
Secondary school	5	3.1	3	2.7	8	2.9
High school	8	4.9	1	0.9	9	3.3
Two-year university programDropping out of school	0	0.0	1	0.9	1	0.4
Intentional	108	66.7	78	70.9	186	68.4	.460^c
Unintentional	54	33.3	32	29.1	86	31.6	.460^c
If it is unintentional, what is the cause? (n:86)
Family did not send to school	17	31.5	14	43.8	31	36.0
Migration	12	22.2	8	25.0	20	23.3
Financial problems	25	46.3	10	31.2	35	40.7
Employment status
Yes	23	14.2	16	14.5	39	14.3	.936^c
No	139	85.8	94	85.5	233	85.7	.936^c
Perceived income
Lower than expenditures	155	95.7	105	95.5	260	95.6	1.00^d
Equal to expenditures	7	4.3	5	4.5	12	4.4
Presence of chronic diseases
Yes	79	48.8	31	28.2	110	40.4	.001^c
No	83	51.2	79	71.8	162	59.6	.001^c
Chronic diseases (n:110)
COLD	3	5.1	0	0.0	4	3.6
Diabetes	37	46.8	21	67.7	58	52.7
Coronary artery disease	4	5.1	1	3.2	5	4.5
Hypertension	68	86.1	28	90.3	96	87.3

Student's t-test; ^bMann-Whitney U test; ^cChi-square test; ^dFisher's Exact test.

*p < .05.

The women with early marriages had a significantly higher number of children: Median = 3, range age at marriage <18 years; Median = 2, range age at marriage >18 years (p = .001). The mean age also differed significantly between the women who married before the age of 18 years old (M = 39.98, SD = 15.1) and those who married after 18 years old (M = 34.21, SD = 10.21) (p < .001). The presence of chronic diseases was significantly different between the women who married before and after 18 years old (p = .001). In fact, 48.8% (n = 79) of the former group and 28.8% (n = 31) of the latter group had a chronic disease (Table 1).

Mental health outcomes

Cronbach alpha coefficients were calculated for the reliability of the BSI and BSI subscales. As shown in Table 2, the subscales of interpersonal sensitivity and psychoticism had low reliability (.40 ≤ α < .60) but the subscales for obsession-compulsion, depression, anxiety, hostility, phobic anxiety, and paranoid ideation had relatively high reliability (.80 ≤ α < l.00). In addition, the subscale somatization was very reliable (.60 ≤ α < .80).

Table 2.

Roma women's scores and reliability for the subscales of the Brief Symptom Inventory.

	Mean ± SD	Min–Max	Cronbach's alpha
Somatization	5.23 ± 5.70	0–28	0.836
Obsession-compulsion	4.92 ± 4.21	0–24	0.679
Interpersonal sensitivity	3.81 ± 2.80	0–16	0.490
Depression	7.51 ± 5.23	0–24	0.785
Anxiety	8.24 ± 5.19	0–24	0.726
Hostility	6.07 ± 4.46	0–20	0.736
Phobic anxiety	2.29 ± 3.14	0–18	0.617
Paranoid ideation	2.93 ± 3.38	0–20	0.639
Psychoticism	2.77 ± 3.04	0–16	0.532
Additional items	3.88 ± 3.78	0–37	0.370
Global Severity Index	0.90 ± 0.57	0.02–3.4	0.939
Positive Symptom Total	20.67 ± 10.04	1–51	0.939
Positive Symptom Distress Index	2.18 ± 0.64	1–4	0.939

Table 3 shows the correlations between the mean scores of the Roma women who married before and after the age of 18 for the BSI subscales, the duration of marriage, and their mother's age at marriage. A significant positive association was found between duration of the marriage and the subscales of interpersonal sensitivity, somatization, depression, anxiety, phobic anxiety, paranoid ideation, psychoticism, and the Global Severity Index in the women who married before the age of 18. Among the women who married after the age of 18, duration of marriage had a significant positive association with the subscales somatization (r = .258) and phobic anxiety (r = .191; p < .001; Table 3). Further, duration of marriage among the Roma women had a significant positive association with all the subscales of the BSI except hostility (Table 3).

Table 3.

Bivariate correlations between Brief Symptom Inventory scores and age at marriage, mothers’ age at marriage and duration of marriage among Roma women.

BSI		Age at marriage <18(n = 162)		Age at marriage ≥18(n = 110)		Total(N = 272)
BSI		Mothers’ age at marriage	Duration of marriage	Mothers’ age at marriage	Duration of marriage	Mothers’ age at marriage	Duration of marriage
Somatization	r	.186	.303	−.074	.258	.075	.296
Somatization	p	.018	< .001	.441	.006	.219	< .001
Obsession-compulsion	r	.203	.127	.088	.188	.147	.152
Obsession-compulsion	p	.010	.106	.363	.050	.015	.012
Interpersonal sensitivity	r	.128	.192	.023	−.043	.073	.123
Interpersonal sensitivity	p	.104	.015	.813	.654	.229	.042
Depression	r	−.110	.338	−.020	.011	−.083	.236
Depression	p	.163	<.001	.833	.907	.173	<.001
Anxiety	r	.039	.249	−.071	.129	−.018	.226
Anxiety	p	.624	.001	.460	.179	.770	<.001
Hostility	r	.132	.057	.027	.120	.084	.085
Hostility	p	.094	.472	.776	.210	.167	.164
Phobic anxiety	r	.184	.233	−.018	.191	.092	.240
Phobic anxiety	p	.019	.003	.855	.045	.132	<.001
Paranoid ideation	r	.189	.207	.033	.140	.108	.174
Paranoid ideation	p	.016	.008	.735	.144	.075	.004
Psychoticism	r	.109	.288	−.009	.021	.051	.219
Psychoticism	p	.166	<.001	.928	.825	.404	<.001
Additional items	r	.010	.265	.068	.302	.015	.290
Additional items	p	.899	.001	.479	.001	.811	<.001
Global Severity Index	r	.137	.324	−.001	.174	.064	.282
Global Severity Index	p	.083	<.001	.989	.069	.289	<.001
Positive Symptom Total	r	.041	.388	−.034	.203	−.013	.358
Positive Symptom Total	p	.605	<.001	.728	.034	.831	<.001
Positive Symptom Distress Index	R	.225	.103	.058	.095	.161	.090
Positive Symptom Distress Index	P	.004	.192	.549	.323	.008	.140

r = Pearson correlation.

*p < .001.

Table 4 presents the results of the multiple linear regression examining the effects of the risk factors (age at marriage, mother's age at marriage, duration of marriage, and the number of children) on the scores for the BSI among the Roma women. Significant predictors of the BSI among the study participants were duration of marriage and somatization, R² = .09, F(4, 267) = 7.56, p = .000, depression, R² = .06, F(4, 267) = 4.92, p = .001, psychoticism, R² = .07, F(4, 267) = 5.90, p = .000, the Global Severity Index, R² = .08, F(4, 267) = 6.93, p = .000, and the Positive Symptom Total, R² = .12, F(4, 267) = 10.51, p = .000 (Table 4).

Table 4.

Multiple linear regression models of effects of risk factors on Brief Symptom Inventory scores of Roma women (N = 272).

	Age at marriage		Duration of marriage		Mothers’ age at marriage		Number of children
	B	p	B	p	B	p	B	p
Somatization	.037	.811	.135	<.001	.258	.106	−.206	.368
Obsession-compulsion	−.038	.746	.046	.032	.326	.007	.002	.992
Interpersonal sensitivity	−.123	.122	.014	.326	.134	.101	.064	.580
Depression	−.129	.374	.093	<.001	−.142	.342	−.292	.172
Anxiety	−.304	.035	.054	.037	.058	.696	.146	.490
Hostility	−.256	.042	−.001	.970	.242	.061	.234	.206
Phobic anxiety	−.092	.290	.046	.003	.183	.040	.044	.729
Paranoid ideation	−.045	.634	.047	.006	.203	.038	−.112	.422
Psychoticism	−.164	.051	.054	<.001	.122	.156	−.286	.021
Additional items	−.053	.607	.081	<.001	.076	.477	−.125	.415
Global Severity Index	−.022	.161	.011	<.001	.028	.088	−.010	.665
Positive Symptom Total	−.473	.078	.241	<.001	.154	.576	−.335	.395
Positive Symptom Distress Index	−.027	.132	.003	.365	.056	.003	.000	.997

Table 5 summarizes the results of separate multiple linear regression analyses with the effects of the mother's age at marriage, duration of marriage, and number of children on the BSI scores in the women who married before and after 18 years of age (Table 5). Among the women who married before the age of 18, longer duration of marriage had a significant positive correlation with somatization, R² = .12, F(3, 158) = 8.48, p = .000, depression, R² = .10, F(3, 158) = 7.25, p = .000, anxiety, R² = .06, F(3, 158) = 4.14, p = .007, phobia, R² = .11, F(3, 158) = 7.93, p = .000, paranoid ideation, R² = .05, F(3, 158) = 3.84, p = .011, and psychoticism, R² = .13, F(3, 158) = 8.79, p = .000 (Table 5). In addition, a significant relationship was found between the duration of the marriage and the Global Severity Index and Positive Symptom Total. Among the women who married after the age of 18, a longer duration of marriage had an effect on somatization, R² = .02, F(3, 106) = 1.58, p = .200 (Table 5).

Table 5.

Multiple linear regression models of effects of risk factors on Brief Symptom Inventory scores among Roma women marrying before and after the age of 18 years.

		Age at marriage < 18(n = 162)		Age at marriage ≥18(n = 110)
Dependent	Independent	B	p	B	p
Somatization	Duration of marriage	.123	<.001	.179	.002	R² = .12F = 8.48p = .000
	Mothers’ age at marriage	.569	.005	−.257	.321
	Number of children	−.039	.887	−.583	.162
Obsession-compulsion	Duration of marriage	.043	.065	.064	.164
	Mothers’ age at marriage	.403	.006	.163	.440
	Number of children	−.077	.701	.179	.596
Interpersonal sensitivity	Duration of marriage	.028	.062	−.011	.721
	Mothers’ age at marriage	.183	.055	.039	.788
	Number of children	.138	.288	−.021	.928
Depression	Duration of marriage	.120	.001	0.041	.453
	Mothers’ age at marriage	−.203	.264	−.053	.834
	Number of children	−.134	.588	−.521	.200
Anxiety	Duration of marriage	.072	.013	.063	.250	R² = .06F = 4.14p = .007
	Mothers’ age at marriage	.152	.408	−.212	.401
	Number of children	.251	.316	.052	.898
Hostility	Duration of marriage	.010	.707	.023	.626
	Mothers’ age at marriage	.287	.076	.041	.853
	Number of children	.195	.375	.417	.237
Phobia	Duration of marriage	.045	.012	.064	.043	R² = .11F = 7.93p = .000
	Mothers’ age at marriage	.309	.007	−.047	.744
	Number of children	.138	.371	−.134	.559
Paranoid ideation	Duration of marriage	.051	.002	.051	.222	R² = .05F = 3.84p = .011
	Mothers’ age at marriage	.276	.008	.045	.816
	Number of children	−.155	.271	.015	.962
Psychoticism	Duration of marriage	.079	<.001	.004	.911	R² = .13F = 8.79p = .000
	Mothers’ age at marriage	.179	.091	−.016	.915
	Number of children	−.378	.009	.040	.864
Additional items	Duration of marriage	.075	.001	.110	.001
	Mothers’ age at marriage	.060	.681	.076	.610
	Number of children	−.063	.753	−.273	.257
Global Severity Index	Duration of marriage	.012	<.001	.011	.083	R² = .02F = 1.58p = .20
	Mothers’ age at marriage	.042	.028	−.004	.886
	Number of children	−.002	.929	−.016	.738
Positive Symptom Total	Duration of marriage	.268	<.001	.215	.037	R² = .02F = 1.58p = .20
	Mothers’ age at marriage	.335	.323	−.237	.613
	Number of children	−243	.599	−.343	.648
Positive Symptom Distress Index	Duration of marriage	.004	.275	.008	.229
	Mothers’ age at marriage	.071	.002	.017	.594
	Number of children	.023	.464	−.039	.435

Discussion

The present study contributes to the growing literature on the mental health effects of early marriage in adult Roma women. More than half of the participants in this study were married early. Among these women, 14.2% married when they were 12–14 years old, and 85.8% married when they were 15–17 years of age (see Table 1 for demographic information). In Turkey, estimates of the prevalence of early marriage among Roma women vary between 30% and 50% (Tanrıverdi et al., 2012; Taylan, 2016).

As the duration of marriage increased, the levels of interpersonal sensitivity, somatization, depression, anxiety, phobia, paranoid thoughts, psychoticism, and scores on the Global Severity Index increased among the Roma women in the early marriage group in this study (Table 3). On the other hand, considerably less psychopathological symptomatology was found in the group that married after 18 years old. The results of the multiple regression analysis supported these findings (Table 5).

In a study undertaken to evaluate the relationship between mental health and age among Roma people in Greece, aside from hostility, women had significantly higher levels of psychological symptoms as measured by the Symptom Checklist-90-Revised (SCL-90-R), and these symptoms had a positive correlation with age (Kotrotsiou et al., 2014). Vorvolakos et al. (2012) compared sociodemographic and clinical features of Roma people with psychological disorders with those of non-Roma people with psychological disorders. They reported that most of the Roma group was female and had married early (mean age of marriage = 16.71 years). The researchers added that the Roma group had more symptoms of major depression, somatization, anxiety, hostility, and phobia than the other group (Vorvolakos et al., 2012). In their study in the United States, Le Strat et al. (2011) found that women who married in their childhood had higher rates of psychiatric and mood disorders rates than those who married in adulthood. Researchers also noted that a higher number of older women in the study had married early (Le Strat et al., 2011).

Marriage can have either positive or negative effects on mental health. It has been suggested that, compared to young adults who have married at a relatively early age, people who marry in early adulthood (20–29 years of age) are more satisfied with their lives and these marriages can be beneficial for their mental health in many respects (Uecker, 2012). Early marriage is a significant stressor that leads women to have difficulties in adolescence and can increase later exposure to other stressors (Alexandridis, 2016; Efevbera et al., 2017; Nour, 2009; Raj, 2010). In the present study, symptoms of psychological distress among Roma women in the early marriage group correlated with the duration of marriage. Longitudinal studies could confirm this and identify the factors that contribute to long-term impacts on mental health.

Limitations

This study examined only one population and the results may have limited generalizability. A similar ethnic group known as “Cono,” who do not consider themselves Roma and live in a different part of Adana, were not included in the study. Additionally, Roma girls younger than 18 years who were married at the time of the data collection were excluded from participating in the study for legal reasons. The study was cross-sectional so causal relationships cannot be confirmed. Lastly, this study was quantitative and did not ascertain potential aspects of the marital trajectory that could influence mental health outcomes. Marriage is a common and traditional practice for Roma women in Roma culture; it makes the conditions of life for Roma women even more challenging due to the difficulties and responsibilities that women face before completing their development in adolescence. A mixed-methods study could be conducted to gather more information about the lived experiences of this population, which is not studied frequently. Additional information could be obtained by interviewing both partners, women and men, in future studies.

Conclusion

This article reports the results of a study of a sample of 272 Roma women in Adana, Turkey. The findings may be specific to this setting but highlight the need for health professionals to protect and promote the mental health of Roma women. Almost two-thirds of women (59.6%) (n = 162) had married before the age of 18 years; of these, 14.2% (n = 22) married before the age of 15. Duration of marriage and age at marriage were significantly associated with higher levels of psychological symptoms in the women who married early. Women who married early reported more psychological symptoms on subscales of interpersonal sensitivity, somatization, depression, anxiety, phobia, paranoid thoughts, psychoticism. Future studies with larger sample sizes that include Roma women in other regions can shed light on the effects of early marriage on the mental state of Roma women.

Compliance with ethical standards

The study was conducted in accordance with the principles of the Declaration of Helsinki. Ethical approval was obtained from the University Ethical Board of Clinical Research and written permission was taken from the city governor. Participation in the study was voluntary and oral and written informed consent were obtained after the women included in the study were given information about the study, confidentiality, privacy, and their right to drop out of the study.

Footnotes

Declaration of conflicting interests

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

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Ayşe İnel Manav

Ayşe İnel Manav, PhD, is an Instructor at Osmaniye Korkut Ata University and an Associate Professor Dr. of Mental Health and Psychiatry Nursing in the Department of Nursing at Osmaniye Korkut Ata University. She studies mental health services for women's mental health, gender inequality, geropsychiatry, and ethnic populations. Her published works focus on mental health services research, conceptual and methodological issues with minority populations, risk behaviors, and disparities in service delivery.

References

Adana Valiliği. (2019). Adana ili/Yüreğir ilçesi . Retrieved October 4, 2019, from https://www.adana.gov.tr/

Afanasieva

Bukrieieva

Glyns’ka

Hlebova

Semikin

(2020). Family roles of Roma and ways to overcome its socio-cultural isolation. Journal of History Culture and Art Research, 9(3), 288–296. https://doi.org/10.7596/taksad.v9i3.2716

Aktepe

Atay

İM

(2017). Child marriages and psychosocial outcomes. Current Approaches in Psychiatry, 9(4), 410–420. https://doi.org/10.18863/PGY.310791

Alexandridis

(2016). Overview of field research findings regarding the educational and health situation of Roma women in Greece. Social Cohesion and Development, 4(2), 173–188. https://doi.org/10.12681/scad.8914

Arthur

Earle

Raub

Vincent

Atabay

Latz

Kranz

Nandi

Heymann

(2018). Child marriage laws around the world: Minimum marriage age, legal exceptions, and gender disparities. Journal of Women, Politics & Policy, 39(1), 51–74. https://doi.org/10.1080/1554477X.2017.1375786

Balkız

Göktepe

(2014). The Roma and their socioeconomic life conditions: A fieldwork carried out in the province of Aydın. Journal of Sociology, 30(1), 1–39. https://dergipark.org.tr/tr/pub/sosder/issue/41005/495511

Bosnjak

Acton

(2013). Virginity and early marriage customs in relation to children’s rights among chergashe Roma from Serbia and Bosnia. The International Journal of Human Rights, 17(5–6), 646–667. https://doi.org/10.1080/13642987.2013.831697

Carrasco-Garrido

Lopez de Andres

Hernandez Barrera

Jimenez-Trujillo

Jimenez-Garcıa

(2010). Health status of Roma women in Spain. European Journal of Public Health, 21(6), 793–798. https://doi.org/10.1093/eurpub/ckq153

Casey

(2014). “Caravan wives” and “decent girls”: Gypsy-traveller women’s perceptions of gender, culture and morality in the North of England. Culture, Health & Sexuality, 16(7), 806–819. https://doi.org/10.1080/13691058.2014.911961

10.

Cochran

W. G.

(1977). Sampling techniques (3rd ed.). John Wiley & Sons.

11.

Coea

Cvorovic

(2017). The health of Romanian Gypsy women in Serbia. Health Care for Women International, 38(4), 409–422. https://doi.org/10.1080/07399332.2017.1292278

12.

Cook

Wayne

G. F.

Valentine

Lessios

Yeh

(2013). Revisiting the evidence on health and health care disparities among the Roma: A systematic review 2003–2012. International Journal of Public Health, 58, 885–911. https://doi.org/10.1007/s00038-013-0518-6

13.

Council of Europe. (2015a). Thematic report by the experts of the CAHROM thematic group on child/early and forced marriages within Roma communities in the context of the promotion of gender equality, CAHROM. (2015) . Retrieved October 2, 2019, from https://rm.coe.int

14.

Council of Europe. (2015b). Child marriage in Albania. Specific view to Roma Community. Observatory for children’s rights. CAHROM. (2015) . Retrieved October 2, 2019, from info@observator.org.al

15.

Cvjetkovic

Jankovic

Bjegovic-Mikanovic

(2017). Differences in predictors of Subjective Well-Being among Roma and Non-Roma young adults in Serbia. European Journal of Public Health, 27, 374–375. https://doi.org/10.1093/eurpub/ckx189.189

16.

Çelik

Yüce Tar

(2015). Roma (gypsy) citizens in Samsun: Living with limited social rights. The Journal of International Social Research, 8(36), 621–635. https://doi.org/10.17719/jisr.2015369528

17.

Danka

(2008). Institutional and legal framework for protecting Roma rights in Turkey. European Roma Rights Center, 17(1), 29–51.

18.

Derogatis

L. R.

(1992). The brief symptom inventory-BSI administration, scoring and procedures manual-II, USA. Clinical Psychometric Research.

19.

Efevbera

Bhabha

Farmer

P. E.

Fink

(2017). Girl child marriage as a risk factor for early childhood development and stunting. Social Science & Medicine, 185, 91–101. https://doi.org/10.1016/j.socscimed.2017.05.027

20.

Foldes

M. E.

Covaci

(2012). Research on Roma health and access to healthcare: State of the art and future challenges. International Journal of Public Health, 57(1), 37–39. https://doi.org/10.1007/s00038-011-0312-2

21.

Gamella

J. F.

(2018). Marriage, gender and transnational migrations in fertility transitions of Romanian Roma women: An ethnographic case study. Intersections-East European Journal of Society and Politics, 4(2), 57–85. https://doi.org/10.17356/ieejsp.v4i2.389

22.

Gaston

C. M.

Misunas

Cappa

(2019). Child marriage among boys: A global overview of available data. Vulnerable Children and Youth Studies, 14(3), 219–228. https://doi.org/10.1080/17450128.2019.1566584

23.

Hajduchová

Urban

(2014). Social determinants of health in the Roma population. Kontakt, 16(1), e39–e43. https://doi.org/10.1016/j.kontakt.2014.01.001

24.

Hotchkiss

D. R.

Godha

Gage

A. J.

Cappa

(2016). Risk factors associated with the practice of child marriage among Roma girls in Serbia. BMC International Health and Human Rights, 16(6), 2–10. https://doi.org/doi10.1186/s12914-016-0081-3

25.

Kotrotsiou

Gourgoulianis

Gouva

Skenteris

(2014). Age and gender as determinants of psychological health for Roma people in Greece. International Journal of Caring Sciences, 7(3), 873.

26.

Le Cook

Progovac

Tran

(2019). Improving the mental health of Roma through research and policies that cross multiple social sectors. International Journal of Public Health, 64, 979–980. https://doi.org/10.1007/s00038-019-01256-7

27.

Le Strat

Dubertret

Foll

B. L.

(2011). Child marriage in the United States and its association with mental health in women. Pediatrics, 128(3), 524–530. https://doi.org/10.1542/peds.2011-0961

28.

Martin

Gamella

(2005). Marriage practices and ethnic differentiation: The case of Spanish Gypsies (1870–2000). The History of The Family, 10, 45–63. https://doi.org/10.1016/j.hisfam.2004.03.004

29.

McFarlane

Nava

Gilroy

Maddoux

(2016). Child brides, forced marriage, and partner violence in America. Obstetrics & Gynecology, 127(4), 706–713. https://doi.org/10.1097/AOG.0000000000001336

30.

McGorrian

Hamid

A. N.

Fitzpatrick

Daly

Malone

Kelleher

(2013). Frequent mental distress (FMD) in Irish Travellers: Discrimination and bereavement negatively influence mental health in the All Ireland Traveller Health Study. Transcultural Psychiatry, 50(4), 559–578. https://doi.org/10.1177/1363461513503016

31.

Medas

(2020). Roma and health: The social determinants of health: Lessons from the case of Roma in Hungary. Journal for Labour and Social Affairs in Eastern Europe, 18(2), 149–173. https://doi.org/10.5771/1435-2869-2020-2-5

32.

Nasrullah

Zakar

(2014). Child marriage and its associations with controlling behaviors and spousal violence against adolescent and young women in Pakistan. Journal of Adolescent Health, 55, 804–809. https://doi.org/10.1016/j.jadohealth.2014.06.013

33.

Nour

N. W.

(2009). Child marriage: A silent health and human rights issue. Reviews in Obstetrics and Gynecology, 2(1), 51–56. PMID: 19399295; PMCID: PMC2672998.

34.

Özkan

A. R.

(2006). Marriage among the Gypsies of Turkey. The Social Science Journal, 43, 461–470. https://doi.org/10.1016/j.soscij.2006.04.003

35.

Pereira

Amaral

Monteiro

Esgalhado

Afonso

R. M.

Loureiro

(2016). Mental health disparities between Roma and NonRoma unemployment insurance beneficiaries in Portugal. Diversidad and Equality in Health and Care, 13(3), 246–250. http://doi.org/10.21767/2049-5471.100058

36.

Raj

(2010). When the mother is a child: The impact of child marriage on the health and human rights of girls. Archives of Disease in Childhood, 95(11), 931–935. https://doi.org/10.1136/adc.2009.178707

37.

Robinson

Oluboyede

Vale

, et al. (2022). Differences in health-related quality of life between the Roma community and the general population in Romania. Journal of Patient-Reported Outcomes, 6, 127. https://doi.org/10.1186/s41687-022-00530-2

38.

Sedlecky

Rašević

(2015). Challenges in sexual and reproductive health of Roma people who live in settlements in Serbia. The European Journal of Contraception & Reproductive Health Care, 20(2), 101–109. https://doi.org/10.3109/13625187.2014.957825

39.

Shives

L. R.

(2008). Basic concepts of psychiatric-mental health nursing (7th ed., pp. 7–32). Lippincott Williams & Wilkins.

40.

Stojanovski

Janevic

Kasapinov

Stamenkovic

Jankovic

(2017). An assessment of Romani women’s autonomy and timing of pregnancy in Serbia and Macedonia. Maternal and Child Health Journal, 21, 1814–1820. https://doi.org/10.1007/s10995-017-2292-1

41.

Svanemyr

Chandra-Mouli

Raj

Travers

Sundaram

(2015). Research priorities on ending child marriage and supporting married girls. Reproductive Health, 12, Article 80. https://doi.org/10.1186/s12978-015-0060-5

42.

Şahin

N. H.

Durak

(1994). Short Symptom Inventory: Adaptation for Turkish youth. Turkish Journal of Psychology, 9(31), 44–56.

43.

Tanrıverdi

Ünivar

Yalçın

Sürer

Açar

Akçay

Hacıoğlu

Yaman

Hastürk

(2012). Evaluate gypsies’ living in Çanakkale according to Purnell’ cultural competence model. Anadolu Hemşirelik and Sağlık Bilimleri Dergisi, 15(4), 244–253.

44.

Taylan

H. H.

(2016). Early marriage research in Roman families of Sakarya. International Journal of Social Science, 52, 221–228. https://doi.org/10.9761/JASSS4875

45.

Turkey Ministry of Family and Social Policies General Directorate of Family and Community Services. (2020). Turkey family structure survey. TMFSPG. Retrieved August 19, 2021, from https://www.ailevecalisma.gov.tr

46.

Uecker

J. E.

(2012). Marriage and mental health among young adults. Journal of Health and Social Behavior, 53(1), 67–83. https://doi.org/10.1177/0022146511419206

47.

United Nations Population Fund. (2012). Marrying too young: End child marriage. UNFPA. Retrieved April 2, 2020, from www.unfpa.org

48.

United Nations Children's Fund. (2012). Child marriage. UNICEF. Retrieved April 2, 2020, from https://www.unicef.org›protection›child-marriage

49.

United Nations Children’s Fund. (2014a). Child protection from violence, exploitation and abuse, child marriage. UNICEF. Retrieved May 1, 2020, from https://www.unicef.org›protection›child-marriage

50.

United Nations Children’s Fund. (2014b). Ending child marriage: Progress and prospects. UNICEF. Retrieved April 2, 2020, from https://www.unicef.org

51.

United Nations Children’s Fund. (2021). The United Nations convention on the rights of the child . Retrieved August 15, 2021, from https://www.unicef.org

52.

United Nations Children’s Fund. (2022). Equality in marriage: Legal age for both men, women should be 18 years. UNICEF. Retrieved March, 2023, from https://www.unicef.org

53.

Van Cleemput

(2018). Health needs of Gypsy travellers. InnovAiT: Education and Inspiration for General Practice, 11(12), 681–688. https://doi.org/10.1177/1755738018774075

54.

WHO. (2013). Global and regional estimates of violence against women: Prevalence and health effects of intimate partner violence and non-partner sexual violence. Retrieved January 2, 2020, from https://www.who.int/reproductivehealth/publications/violence/9789241564625/en/

55.

WHO. (2016). Child, early and forced marriage legislation in 37 Asia-Pacific countries. Retrieved January 2, 2020, from http://archive.ipu.org/pdf/publications/child-marriage-en.pdf.

56.

Vorvolakos

Samakouri

Tripsianis

Tsatalmpasidou

Arvaniti

Terzoudi

Livaditis

(2012). Sociodemographic and clinical characteristics of Roma and non-Roma psychiatric outpatients in Greece. Ethnicity & Health, 17, 1–2, 161–169. https://doi.org/Z0.1080/13557858.2012.654767

57.

Voicu

Popescu

(2009). ‘Roma women, known and unknown. Family Life and The Position Of The Woman In The Roma Communities – Research Report’, 2009.