This was a cross-sectional study using a non-probabilistic sequential sample of 150 candidates who voluntarily sought to terminate pregnancies, in a public health service, in Lisbon, between September 2008 and June 2009. The research project was evaluated and approved by the institution's Ethics Committee.

The data collection was carried out in the Unwanted Pregnancy Consultation unit at maternity Dr. Alfredo da Costa, and it consisted of a single evaluation done during the pre-consultation period (first appointment before abortion). After explaining the study and guaranteeing the confidentiality and anonymity of the evaluation, the participants granted their informed consent.

The instruments were socio demographic questionnaire, tailored for the study in order to characterize the study sample on a number of socio-demographic variables (including age, nationality, religion, ethnicity, educational level, professional situation, socio-economic level, and living conditions).

Depression Anxiety Stress Scales (DASS) (24), Portuguese adaptation (25) − this is a 21-item scale, divided in three subscales of depression, anxiety and stress, each with 7 items. The responses are given in a 4-point scale, where the respondent presents the level of agreement with the item (e.g. statements regarding negative emotional states felt during the past week). The score for each subscale is obtained by the sum of the items, with the minimum result of 0 and the maximum of 21 (the higher the score the more negative the affective state). The Portuguese modified version presented a good internal consistency (α ranged between 0.74 to 0.85), presenting similar psychometric characteristics for the Portuguese population as the original scale (25).

Satisfaction with Social Support Scale (SSSS) (26) − this 15-item scale measures four dimensions of social support: satisfaction with friends (5 items), intimacy (4 items), satisfaction with family (3 items), and social activities (3 items). The responses are given in a 5-point Likert scale, and results vary between 15 and 75 points. The Portuguese version presented a good internal consistency (α=0.85), (26).

Emotional Assessment Scale (EAS) (27), Portuguese adaptation (28) − it is a 24-item scale that evaluates the emotional reactivity of fundamental emotions: fear, happiness, anxiety, guilt, anger and surprise. The responses are given using a 10cm visual analogue scale, which ranges from “least possible” on the left to “most possible” on the right. The scoring of the intensity reported regarding each emotion word is obtained by measuring the distance (in millimeters) between the left endpoint to the slash mark made by the respondent. In the Portuguese study, the item's internal consistency was considered good (α values ranged between 0.7286 to 0.8818), (28).

Decision Conflict Scale (DCS) (29), Portuguese adaptation (30) – this is a 16-item scale that assesses the perceived difficulty in the decision-making process: uncertainty in choosing among several options, contributing factors and quality of the decision-making process. Responses are given on a 5-point Likert scale, which ranges from “completely disagree” to “completely agree”. The score is obtained by calculating the mean of the responses to each item. The higher the final score the greater the difficulty in making the decision. The Portuguese validation study presented an excellent internal consistency (Total α=0.9024) (30).

Beliefs and Values Questionnaire (BVQ) about sexuality, developed by the researchers and tailored for the Portuguese population (31) – this is a 38-item scale that evaluates the values and beliefs on sexuality in the dimensions of 1) Maternity (which assesses the perception of motherhood as a central project of being a woman); 2) Reproduction (which assesses the extent to which female sexuality is seen essentially in its procreative function); 3) Affectivity (assessing to which extent female sexuality is perceived as an affective area); 4) Abortion (which evaluates the negative representations of abortion); 5) Pleasure (which assesses the perception of female sexuality as a pleasure area). The responses are given using a 5-point Likert scale, ranging from “I totally disagree” to “I totally agree”. The final score is obtained through the calculation of the mean of the responses to the items. The validation study presented acceptable α values (0.560 to 0.782) (31).

Student's t-tests were performed to compare two means, and multivariate analysis of variance (MANOVA) followed by univariate analysis of variance (ANOVA) were used to compare more than two sets of means. When homoscedasticity requirement was not met in the ANOVA, we used Welch's corrected ANOVA. Tukey's post-hoc tests were used when ANOVA proved to be significant to identify the groups with statistically significant mean differences. Finally, to analyze the association pattern between the variables of the subscales, we used a Categorical Principal Component Analysis (CATPCA). Statistically significant effects were accepted for p<0.05.

Data analysis was conducted using the statistical software SPSS (Version 19, SPSS Inc., Chicago, ILL, USA).

The study sample was composed of 150 pregnant women, until tenth week of pregnancy, who made the decision to have an abortion.

Information on the participants’ clinical and abortion history is presented in Table 1.

The participants' age ranged from 13 to 44 years (M=26.03, SD=6.987), 76% were Portuguese and 20% from African countries; 73.3% were Caucasian and 26.7% were Black. Most participants had a secondary educational level (45.7%) and were professionally active (61.3%), the remaining 27.3% were students, and had a low socio-economic level (51.3%).

Analyzing psychological adjustment to the decision-making process, the means of all the three subscales including depression, anxiety and stress were low (M_depression=5.9; M_anxiety=5 and M_stress= 8.7), the values being below the considered “risk zone”, indicated that the women in our sample were psychologically adjusted to the condition.

As for Satisfaction with Social Support (SSSS), the means were high in all subscales, indicating a high perceived social support. Moreover, the means reached 75% of the maximum score in all subscales, except for the “social activities” subscale, i.e, the participants were less satisfied with their social activities than with any other social support dimension.

Concerning the Emotional Reactivity (EAS), we observed that the emotion with the highest mean was “sadness” (M=62.2), followed by “fear” (M= 61.7), “anxiety” (M=45.7), “surprise” (M=43.9), “anger” (M=41.5), “guilt” (M=35.1) and, finally, “happines” (M=18.0).

Moreover, the Conflict with the Decision (DCS) presented low mean score (M=1.56), indicating a low conflict in the decision-making. Analyzing some of the items individually, we observed that the first item ("this decision was easy to make") had the highest mean (M=3.20), i.e., despite an overall low conflict, women considered this a somewhat difficult decision, but not extremely difficult. However, in the second item ("I'm sure of the decision to make"), (M=1.53), the participants showed that, despite the difficulty, they were sure about their decision.

Concerning the values and beliefs about sexuality, pregnancy, maternity and abortion (BVQ) of the sample, which was originally used to assess BVQ, statistically significant differences with the general population for the subscales “motherhood” (p=0.017) “reproduction” ( p=0.026) and “affection” (p=0.047) were found. Since all the test statistics were positive (our sample vs. general population), we could infer that the women from our sample privileged sexuality as a source of affection, they had stronger beliefs related to the idea of motherhood as a central project in a woman's life, as well as regarding reproduction as a primary function of female sexuality, and had a more positive view of abortion.

We also analyzed individual and relational variables considered relevant in the literature. Of these, we presented only the findings that had statistically significant effects (p<0.05).

Age group was one of the most important socio-demographic variables to explain the beliefs and values regarding sexuality (Pillai's trace=0.315; F(20,576)=2.458; p<0.001; η² _p=0.079; π=0.998). The subsequent ANOVA revealed statistically significant effects of age on the dimensions “reproduction” (F(4,145)=3.916; p=0.005; η² _p=0.098; π=0.72) and “Pleasure” (F(4,145)=2.598; p= 0.039; η² _p=0.067; π=0.72).

Regarding the age groups, teenagers (13 to 18 years old), had the strongest beliefs of reproduction as the primary function of female sexuality (13 to 18 and 19 to 24, p=0.04; 13 to 18 and 25 to 29, p=0.003; 13 to 18 and 30 to 35, p=0.021), (Table 2).

The relational context was influenced by beliefs and values (Pillai's trace=0.120; F(10,288)=1.833; p=0.05; η² _p=0.06; π=0.842). Subsequent analysis revealed significant effects over the dimensions “Maternity” (F(2,147)=3.442; p=0.035; η² _p=0.045; π=0.638) and “Pleasure” (F(2,147)=3.576; p=0.03 η² _p=0.046; π=0.656). Moreover, women who were married or cohabiting tended to assess motherhood as the central project of womanhood, and female sexuality as an area of pleasure (p=0.030), in opposition to women who were dating or had casual relations (p=0.024).

Furthermore, stress and anxiety levels were marginally affected by the socio-economic level (Pillai's trace=0.107; F(9,438)=1.796; p=0.06; η² _p=0.036; π=0.806). A post-hoc analysis was performed and showed significantly higher levels of anxiety (p=0.005) and stress (p=0.035) in women with lower socio-economic levels when compared to those with higher levels. The socio-economic level also influenced the perceived social support dimensions (Pillai's trace=0.145; F (12,435)=1.836; p=0.04; η² _p=0.048; π=0.895). Also, we found that women with lower socio-economic levels were more unsatisfied regarding the intimate social support (p=0.034), and had stronger beliefs regarding motherhood (p=0.003) and procreation¹ (p=0.027) as the women's main functions.

Moreover, the level of overall satisfaction (with social support and social activities) was higher among participants with a higher social status, and the dimension of sexual pleasure was more valued by these women (p=0.037), (Table 3).

Also participants with a lower education level felt more anxious, had less satisfaction in perceived social support and the level of intimacy , valued the project of motherhood more, and viewed sexuality as an area of affection (p<0.05).

On the other hand, participants with higher educational levels tended to privilege pleasure in their sexual experience (Table 4). Moreover, contraception use had significant effects on women's beliefs and values towards sexuality. Women who did not use contraception had stronger beliefs about reproduction as the primary objective of sexual intercourse, than women who used contraception (F (1,148)=3.635; p=0.009; η² _p=0.046; π= 0.752).

Levels of stress, anxiety and depression differed significantly among women with different gestation periods (Pillai's trace=0.052; F(3,146)= 2.657; p=0.05; η² _p=0.042; π=0.711). A post-hoc analysis revealed that women with shorter gestation periods exhibited lower levels of depression (p=0.01), anxiety (p=0.012) and stress (p=0.034), less conflict in the decision-making (p=0.003), and less firm beliefs about sexuality as a way to find affection (p=0.023).

The number of previous pregnancies was also influential on women's beliefs and values regarding sexuality (Pillai's trace=0.325; F(30,725)= 1.656; p=0.016; η² _p=0.065; π=0.993). Thus, primiparous women that were more fearful (p= 0.014) had stronger beliefs about the importance of motherhood in a woman's life (p<0.001), about pleasure in female sexuality (p=0.007), and had more negative feelings regarding abortion (p=.043), when compared with women with two or more previous pregnancies. Likewise, nulliparous women were also fearful (p=0.038), but they had weaker beliefs regarding motherhood as the central project of womanhood (p=0.001), and about the experience of sexuality as a source of pleasure (p=0.004).

No statistically significant differences were found regarding religion, ethnicity, or previous abortions at the level of psychopathological symptoms, satisfaction with social support, emotional reactivity, or conflict in the decision-making, values or beliefs.

To better understand and characterize the psychological adjustment of the participants, when faced with the decision to have an abortion, we used a Principal Component Analysis with the objective of analyzing the association between several subscales (Figure 1).

The results showed that the variables depression, anxiety, stress, fear, happiness, guilt, anger, surprise, sadness, conflict in the decision-making and reproduction were opposite to satisfaction with friends, intimacy, satisfaction with family, social activities, abortion and pleasure. Thus, women with higher scores in satisfaction with social support had lower psychopathological symptoms, lower emotional reactivity and lower conflict with the decision-making.

Regarding values and beliefs, we noticed that the highest scores of the variables abortion and pleasure, were opposite to those of reproduction, i.e. women with a higher representation of sexuality as a pleasurable action and a poorer representation of abortion, had weaker beliefs about the procreation as the major function of female sexuality.

On the other hand, the variables motherhood and affectivity were orthogonal to the remaining variables, i.e. the perception of motherhood as a determinant project of the female status, as well as the perception of sexuality as an area of affection, did not relate to the remaining variables.