Maternal Chronic Depression Affects Love Styles: A Cohort Study in Southern Brazil

Given the importance that love has in individuals’ lives, and considering that more specific studies about the relationship between depression and the way people love are lacking, the aim of the current study was to evaluate the effect of major depressive disorder (MDD) on “love styles.” This study was nested in a cohort of adolescent mothers. Women were assessed for MDD at each point of assessment: during pregnancy (T1), at between 30 and 60 postpartum days (T2), and at approximately 30 postpartum months (T3), by using the Mini International Neuropsychiatric Interview (MINI PLUS) tool. The Love Attitudes Scale (LAS) was administered to assess the six love styles at T3. Logistic regression analyses were carried out to express the strength of depression on love styles. We found that chronic major depression had the largest effect on the love styles (low Eros, high Mania, and high Agape), followed by current major depression. However, the same association was not observed in the perinatal period. Thus, the assessment of love style is relevant as it seems that an MDD-related injury in maternal life is related to their attitudes about love and there could be possible harmful consequences of these attitudes to the mother and the whole family.

Although MDD has been associated with damage in interpersonal relationships, there is still a lack of evidence in the literature regarding the association of MDD, and mainly, its chronicity, with love styles. Considering the role that love has in an individuals' life, the fact that love styles are changeable, and major depressive episodes (which could bring harm to interpersonal and romantic relationships), the present study aimed to verify the effect of major depressive episodes on the love styles of young mothers, from the gestational period to 30 months of life of their children. The hypothesis tested in this study is that women diagnosed with MDD during the study -and especially those who presented chronically -tend to exhibit love styles pathologically or in an undesirable and unsuitable way.

Method Study Design
This study was conducted with a cohort of adolescent mothers in the urban area of Pelotas, a city located in the Brazilian state of Rio Grande do Sul. The overall study intends to evaluate and follow the mental health of adolescent mothers and the overall development of their child.

Participants
Between October 2009 and March 2011, the participants were recruited from 47 primary healthcare units and three obstetric ambulatory care units that are part of the National Public Health System. All pregnant adolescents (15 to 19 years old) who were enrolled in a prenatal program of the municipal health department were asked to participate in the study. At the time of the present study, the sample was calculated according to prevalence of MDD in pregnant adolescents and its impact on the development of the child, resulting in a total of 544 participants. Figure 1 shows the diagram flow of this study.

Measures Major Depressive Disorder
A validated Portuguese version of the Mini International Neuropsychiatric Interview (MINI PLUS; Amorim, 2000) was used during pregnancy (Time 1 -T1); between 30 and 60 days postpartum (Time 2 -T2); and at approximately 30 months postpartum (Time 3 -T3). The instrument comprises a short structured interview with adequate validity and reliability to assess current MDD according to DSM-IV and ICD-10 criteria in less time than other diagnostic interviews, intended for use in clinical practice and research. The interviews were Böhm, de Avila Quevedo, da Cunha Coelho et al. conducted by previously trained psychologists. The MDD variables were categorized as: "never" (included mothers who had not been diagnosed with major depressive disorder at any time during the study -neither in T1 nor T2 nor T3); "perinatal" -period defined as from conception to 1 year after birth (included mothers who were diagnosed with major depressive disorder during pregnancy -T1 -and or in the postpartum period -T2); "current" (mothers who presented major depressive disorder in the final stage of the study -T3) and "chronic" (mothers who were diagnosed with major depressive disorder in the perinatal and current period -T1 and/or T2 and T3).

Love Styles
A Brazilian validated version of the LAS (Berti et al., 2011) was administered. The used scale build by Hendrick et al. (1998), assesses the six love styles suggested by Lee (1977) and comprises 24 questions, separated into six corresponding subscales, with answers ranging from 1 to 5. Cronbach's alpha reliabilities for all six subscales in the Brazilian sample were generically acceptable: Eros (.741), Storge (.898), Pragma (.648), Mania (.876), and Agape (.881) with the exception of Ludus (.448).
Analyses were conducted by determining the lowest or highest tertile of each love style that hypothetically embodies the "pathologic lovers," based on the characteristics of each love style and on prior research. Love style variables were transformed in dichotomous variables in which one-third of every variable represents the most pathological or undesirable pattern of love as follows: low Eros (low sexual attraction, low complicity, and low emotional intensity); high Ludus (very low commitment, high game-playing, and big lack of self disclosure); low Storge (low friendship); high Pragma (very low emotional, too selfish); high Mania (very obsessive, selfcentering and impulsive, high humor fluctuation, and high agony feelings associated with pathological love); and high Agape (too self regardless, low attention in self needs, and high level of sacrifice).
Participants answered the questions, taking into account their current or last romantic relationship.

Socio-Demographic Data
A self-report questionnaire was used to obtain data related to age, marital status, education, and socioeconomic status. Age was dichotomized into adolescent (ages 15 to 19) according to the World Health Organization (WHO; WHO, 2014), and emerging adulthood (ages 20 to 23) terminology borrowed from Arnett (2000). Marital status was categorized as living or not with a partner. Education (years of study) was grouped into four categories: less than 5 years, between 5 and 8 years, between 8 and 11 years, and between 11 and 14 years. Statistical analyses were conducted using simple frequencies to characterize the sample and demonstrate the distribution of major depressive episodes as well as love styles. Bivariate analysis was used to establish associations between major depressive episodes, other variables of interest, and love styles. Further, logistic regression analyses were carried out to express the strength of depression over love styles, controlling for possible confounders. SPSS (Version 21.0) for Windows was used for the statistical analysis.

Procedure
After the identification of the potential participants from the healthcare units, pregnant adolescents and their parents were contacted via telephone. In case of agreement to participate in the study, an appointment was made for a home interview. Then, a written informed consent was obtained, and if they were younger than 18 years old, permission and informed consent was requested from their parents.
Three interviews were conducted: during pregnancy (T1), with a mean gestational age of 23 weeks; between 30 and 60 days postpartum (T2); and at approximately 30 months postpartum (T3). T1 and T2 interviews were conducted in home visits, when instruments for the mental health assessment (including the MINI PLUS) and socio-demographic data were collected. The present study (T3) was conducted in the Psychological Clinic of the Catholic University of Pelotas, when -in addition to the tools previously used -the LAS was also administered.

Results
Of the initial 544 young mothers with a 30-month post-partum follow-up, 540 (99%) completed the demographics questions, 502 (93%) completed the diagnostic assessment for MDD, and the missings of the love styles variables is indicated in the Table 1. Their mean age was 20.1 years (SD ± 1.5); 39% had studied for between 5 and 8 years; 69% were inhered to C socioeconomic status; and 55% were living with a partner. The prevalence of MDD in the perinatal, current, and chronic periods were 10%, 19%, and 13%, respectively (Table 1).
In a bivariate analysis, low Eros was associated with socioeconomic status D/E (p = .009), living with a partner (p < .001), and the presence of major depressive episode, in the current period and chronically (p = .003).
high Ludus (p = .004) and low Storge (p = .014) were related to not living with a partner. High Pragma had an association with low education (p = .019), while high Mania was associated with low education (p < .001), and major depressive episodes chronicity (p ≤ .001). High Agape was associated with living with a partner (p = .001), and there was a presence of major depressive episodes in the perinatal, current, and chronic periods of the episodes (p = .038; Table 2).
After adjusting for possible confounding variables, low Eros was still related with socioeconomic status, not living with a partner, and the presence of current and chronic major depressive episodes. Mothers from socioeconomic status D/E were 7.9 times (95% CI [3.1, 20.6]) more likely to present low Eros when compared with mothers from socioeconomic status A/B. Women who did not live with their partner were 4.9 times (95% CI [3.0, 7.9]) more likely to show low Eros in relation to those who did. Those who were diagnosed with current major depressive episodes were 2.1 times (95% CI [1.2, 3.8]) more likely to be in the low Eros group than women who had never been diagnosed with major depressive episodes; those who presented chronic episodes were 2.6 times (95% CI [1.3, 5.1]) more likely to show low Eros than those who never had any major depressive episodes.
Ludus and Storge remained associated with not living with a partner (p = .004 and p = .011, respectively).
Mothers who were not living with a partner were 1.7 times (95% CI [1.2, 2.5]) more likely to exhibit high Ludus and 1.6 times (95% CI [1.1, 2.3]) more likely to present low Storge.  Mania maintained an association with age, education, and the presence of depressive episodes. The youngest mothers were 1.4 times (95% CI [1.0, 2.0]) more likely to disclose high Mania in relation to the oldest mothers.
Regarding education, women who studied for few years were 4.8 times (95% CI [1.8, 12.8]) more likely to present high Mania than those who had a higher education. Moreover, mothers with chronic depressive episodes were 2.7 times (95% CI [1.4, 5.1]) more likely to show high Mania when compared to mothers who had never had a major depressive episode.
Finally, with respect to Agape, mothers who were currently experiencing a depressive episode were 1.8 times (95% CI [1.1, 3.2]) more likely to present high Agape. Those mothers who suffered from chronic depressive episodes were 1.9 times (95% CI [1.1, 3.2]) more likely to show high Agape in comparison to those who have never had a major depressive episode (Table 3).

Discussion
This study aimed to verify the effect of major depressive episodes on the romantic love styles of emerging adult mothers. Although the literature suggests that MDD impairs romantic relationships (Burke, 2003;Vujeva & Furman, 2011), this cannot be affirmed in the present study because love styles were assessed only at the T3 stage. Therefore, appropriate statistical treatment of the data was applied and does not necessarily point to a causal relation between depression and love styles. Considering this, an inverse relationship is also possible in which the love styles trigger the depression.
We found that chronic major depression has the largest effect on love styles (low Eros, high Mania, and high Agape), followed by current major depression. However, the same relationship was not observed in the perinatal period, probably because women are more focused on the baby and the transition to motherhood during this period than on their romantic relationship. Moreover, perinatal depression seems to be a very specific form of depression, probably associated with changes in ovarian hormones, even though the underlying mechanisms need further explanations (Albert, 2015;Bloch, Daly, & Rubinow, 2003).
Eros can be considered the healthiest love style. As explained by Levine, Aune, and Park (2006), people with high Eros scores may be able to maintain their relationships more easily and for longer if they increase contact, show tokens of affection, and demonstrate behavioral adaptations. Besides being characterized by strong physical attraction, high complicity, and emotional intensity (Hendrick & Hendrick, 1986), Eros lovers commonly Chronic Depression and Love Styles conduct their personal relationships in a warm and positive way and are shown to be content about life and work and show self-confidence in love (Lee, 1977). Thus, it is not surprising that low Eros scores matched with major depressive states, considering the common characteristics and the interpersonal processes on MDD such as low interest or pleasure from social interactions, isolation, extreme feelings of worthlessness or guilt (Hames et al., 2013).
On the opposite side of this are Mania and Agape, already found to be associated with pathological love in another study (Sophia et al., 2009) and associated with depression in the present research. Studies found that people with MDD tend to engage in excessive interpersonal feedback seeking behaviors, core beliefs reflecting insecurity in relationships, interpersonal dependency, and insecure attachment styles (Hames et al., 2013). Characteristics of a manic lover seem to present a similar direction, including loneliness, lack of friends, and lack of enjoyable work (Lee, 1977). Moreover, manic lovers are more likely to enact negative relational maintenance behaviors (Goodboy, Myers, & Members of Investigating, 2010). This pattern can, in turn, possibly play an important role in the maintenance of the MDD, providing a feedback cycle of the disorder.
An important characteristic of MDD is the difficultly that the individual experiences in having positive attitudes toward themselves, as it explains part of Beck's (Beck, Rush, Shaw, & Emery, 1979) cognitive triad, also known as the negative triad, that is a cognitive-therapeutic view of the three key elements of a person's belief system present in depression (negative view of oneself, environment and future; Pössel & Pittard, 2016). High Agape scores (too self regardless, low attention to self needs, and high level of sacrifice) seem to be a symptom of MDD. It is likely that the focus on their lover's needs is used to take the focuses off their own suffering and needs. Furthermore, making sacrifices for their lover can dissemble worthlessness feelings. Thus, some damages to interpersonal and loving relationships related to the presence of MDD are evident.
The finding that mothers who were not living with a partner interestingly presented low Eros, high Ludus, and low Storge. Assuming that Eros and Storge are defined by commitment in relationships, and Ludus is defined by avoiding commitment, it seems this was a group (low Eros, high Ludus, and low Storge) for who is difficult to establish intimate relationships and maintain a partnership.
Another finding was that women who were inhered of low socioeconomic status presented low Eros. One possible explanation for this is that Eros-style love is very intense and the relationship has a very important place in the individuals' life, and financial difficulties can cause problems in personal relationships, particularly this love style, which demands a considering level of surrender.
Finally, it was verified that adolescent mothers presented high Mania, which can be related to the "storm and stress" period when unstable mood is common (Arnett, 1999). This association was not found in another study (Shulman & Scharf, 2000), which identified Storge as more prevalent in the youngest group. Considering that romantic relationships can positively affect development and also place adolescents at risk of problems (Furman, 2002), the results call attention to the need to take their "love difficulties" seriously, as a potential risk for present and future mental health.
The results of this research need to be interpreted in the light of some limitations. First, as already mentioned, love style information was available only in the last period of the study. Evaluation of love styles at other points in the research could enrich analyses and ground a causal relation between variables. Another limitation was the lack of data regarding relationship characteristics, such as duration of the relationship, level of satisfaction, Böhm, de Avila Quevedo, da Cunha Coelho et al.

Conclusions
Despite the limitations, this study contributes to the literature on mental health and romantic relationships, as the current literature is scarce. The results point to the importance of MDD, and especially its chronicity, in the relationship with love styles. Furthermore, this study extended the notion, observed in previous research (Gadassi et al., 2011;Zeigler-Hill, Britton, Holden, & Besser, 2015), that people without positive attitudes toward themselves and their close relationships may have more pathological and unsuitable love interactions.
It seems that another injury related to MDD in maternal life impacts the way that the mothers act about love, and the consequences that a pathological pattern can bring to the relationship with the whole family. Thus, it is important to identify the love style in woman suffering from MDD, given the importance that love relationships have in an individual's life, and even more so in a young mother's life.
The healthier a woman's love, the healthier the interaction with her partner and, consequently, with her baby.
Further, the healthier her interpersonal relations, the better her prognosis regarding major depression remission.

Funding
This research supprted by Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul (FAPERGS).