Introduction

Maternal-fetal attachment describes the relationship between mother and her fetus and is an emotion toward the unborn child, which starts from the beginning of pregnancy and reaches its peak during the second and third trimesters of pregnancy (1, 2). This feeling is influenced by maternal age, educational level, socioeconomic status, marital status, gestational age, parity, mothers mental imagine of herself, obstetric and medical problems during pregnancy, social support, maternal anxiety and psychiatric status, accepting the pregnancy, and unwanted pregnancy (3, 4).

Anxiety in pregnant women, especially in nulliparous women who are not ready for motherhood, reduces maternal-fetal attachment and mother's ability to assume maternal role (5). In a study by Ossa et al. (2011), maternal-fetal attachment was significantly higher in women with intended pregnancy compared to those with unintended pregnancy (6). In addition, in a study by Ustunsoz et al. (2010), the maternal-fetal attachment score was higher in planned pregnancies than unplanned ones (7). However, in a study by Torshizi et al. (2012) performed in Birjand, mean score of maternal-fetal attachment was not significantly different between wanted and unwanted pregnancies (8).

Maternal-fetal attachment plays an important role in accepting maternal role, child's growth and development, increasing tendency to breast-feeding (9), promoting mothers' confidence in playing maternal role, and causing the mother to show more compatibility with postpartum period and give positive response to infant's behaviors (4, 10). In addition, mothers with higher fetal attachment have stronger interaction with their infant (4). Mothers with weak attachment to their fetus reported high levels of anxiety and depression that can lead to adverse pregnancy outcomes (8).

The results obtained by Kelly (2001) revealed that maternal-fetal attachment is directly associated with health training (11). Belleini et al. (2007) in a study showed that training courses during pregnancy have a positive impact on maternal-fetal attachment (12). Ample studies have been performed so far in the field of maternal-fetal attachment including training of attachment behavior to fetus, touching the fetus, counting fetus’s movements, and relaxation training (3, 4, 10, 13, 14).

Relaxation is a coping strategy for stress, and in behavioral and alternative therapies and simple medical treatments, relaxation and maternal-fetal attachment are practiced after a brief training program (10, 15, 16). There are various techniques for relaxation including guided imagery, hypnosis, massage, meditation, muscular progressive relaxation, and yoga (15). Some studies have shown that guided imagery leads to relaxation, reduced anxiety, and fetal attachment (16-19).

Guided imagery during pregnancy can improve the feeling of mothers toward their fetus and help them comply with their role (10). Deutsch et al. reported that women who imagine themselves as able mothers during pregnancy had more compliance with maternal role and were more competent in their maternal role during the postpartum period (20).

Guided imagery as a mind-body technique is based on the assumption that the mind and body are linked and reinforce each other (21). This method can be learned by an instructor or self-training books or tapes and can be used at times of pain or anxiety and mental stress (22). Zahourek ​​(1988) suggests that guided imagery affects physiologic and psychological status and behaviors of a person, promotes health, reduces stress, and enhances positive feelings (19). Rogers (1983) believes that humans and environment are energy fields, which are continuously interacting, such that change in one leads to transformations in the other. Women form an attachment with fetus during pregnancy if their energy and environment are directed toward this goal. Attachment behavior is a symbol of human-environment interaction and is a process with energy context (23).

Maternal-fetal attachment, which is lower in unplanned pregnancies, is of great significance for desirable pregnancy outcomes. Guided imagery triggers formation of attachment behaviors (23); however, different psychological aspects of guided imagery have not been evaluated yet. Thus, this study was performed to determine the effect of guided imagery on maternal-fetal attachment in nulliparous women with unplanned pregnancy in Mashhad health centers.

 Materials and Methods

This study is a clinical trial with a pretest-posttest design that was performed on 67 nulliparous women (35 in the intervention and 32 in the control groups) with unplanned pregnancy and gestational age of 33 weeks, who were referred to 10 health centers of Mashhad, Iran, during August-October 2014. Based on the results of a pilot study and the formula of comparison of means and 99% confidence coefficient and power of 90%, sample size was calculated to be 35 cases in each group.

In the control group, three cases (two due to failure to complete the questionnaire after delivery and one due to lack of desire to continue participating in the study) were excluded from the study. The inclusion criteria were nulliparous women, obtaining a score of 0-3 from London questionnaire, singleton pregnancy, gestational age of 33 weeks, and living with husband. The exclusion criteria were preterm labor, obstetric problems during pregnancy, infertility history, and education in Medical Sciences and Psychology.

Data collection tools included personal-family information form, London questionnaire to assess unplanned pregnancy, depression, anxiety, and stress questionnaire (DASS 21), and Cranley's maternal-fetal attachment questionnaire. Support of husband and others (father, mother, sister, brother, relatives, friends, and colleagues) was assessed using two questions in the demographic form, such that the subjects selected a number from zero to 10 with 0 indicating lack of support, 5 showing average support, and 10 exhibiting complete support.

London questionnaire comprises of six questions investigating the use of contraceptive methods, timing of pregnancy, intention and desire for pregnancy, communication with the sexual partner, and preparation for pregnancy. All the questions are scored from zero to two with total scores ranging between zero and 12. The scores of 0-3 demonstrate unplanned pregnancy (24).

Cranley's questionnaire includes 24 items, which can be divided into five sub-groups of interacting with fetus, mother’s differentiation between herself and fetus, accepting maternal role, attributing some characteristics to the fetus, and devotion. The questionnaire was rated using a 5-point Likert-type scale (certainly no [1], no [2], do not know [3], yes [4], and definitely yes [5]). The minimum and maximum scores were 24 and 120 with higher scores representing stronger attachment (4).

Dass 21 questionnaire comprises of 21 questions evaluating depression, anxiety, and stress. The scale is rated using a 4-point Likert scale with responses ranging from never to very high (zero to three). Scores of 6-8, 4-5, 6-5 exhibit mild depression, anxiety and stress, respectively, and scores of > 10, 14, 17 show severe depression, anxiety, and stress (25).

Content validity of the personal-family information form and London questionnaire was confirmed and reliabilities of London, Cranley's, and Dass 21 questionnaires were confirmed with alpha coefficients of 0.71, 0.76, and 0.90, respectively. The researcher adjusted the content of guided imagery in relation to maternal role with cooperation of a consultant; after the content was approved by the consultant, the participants were referred to the studio of recording audio and then recorded training CD of maternal role imagery. After final approval of the guided imagery CD by the consultant, it was given to the subjects. Social and economic levels of the subjects were determined with respect to their job and educational level.

After obtaining approval of the Ethics Committee of Mashhad University of Medical Sciences, sampling was performed through combined, stratified, cluster, and available sampling methods. Five health centers of Mashhad (health centers of No 1, 2, 3, Samen, and 5) were considered as one class (five classes in general). From the list of existing centers in that area, two centers were randomly selected as a cluster. Random selection was in a way that both names of two centers were written on two separate sheets and placed into a draw bag, the first selected sheet was assigned to the intervention group and second sheet to the control group. Given the number of classes, a total of 10 health centers were selected.

If nulliparous women who met the inclusion criteria had unplanned pregnancy, the researcher explained the aims of the study to them; London questionnaire was given to them and if they obtained a score of 0 to 3, other inclusion criteria were evaluated and if they met all the criteria to enter the study, informed consent form was completed by them. The researcher reminded the participants the appointment through phone calls on the day before the intervention. Then, in the 34^th week of pregnancy, the participants in both groups completed the personal-family form and Cranley's and DASS 21 questionnaires.

The experimental group received one session of guided imagery regarding maternal role in 34^th week of gestation in groups of four and seven. Guided imagery was taught through training classes for the mothers at health centers that had comfortable chairs and were sufficiently quiet. The techniques and the effects of guided imagery were taught to the subjects. Afterwards, the mothers were asked to take a few deep breaths, switch off their phones, and focus their attention on the CD of maternal role imagery and imagine their maternal role throughout the mental imagery, which lasted for 20 minutes.

Thereafter, the researcher gave the guided imagery CDs to the mothers and they were asked to perform guided imagery at home twice a week during two weeks. The researcher followed the subjects through phone calls in terms of performing guided imagery. The control group only received the routine care. Finally, the participants completed the Cranley's questionnaire in the 36^th week of pregnancy. Pre-test and post-test results were compared between the two groups.

To analyze the data, descriptive statistics (mean, standard deviation, median, interquartile range, frequency distribution) and independent t-test, paired t-test, Chi-squared, Fisher’s exact test, and Mann-Whitney U tests were run using SPSS version 21. P-value less than 0.05 was considered statistically significant.

 Results

The two groups were matched for age, husband and others' support, such that mean ages in the intervention and control groups were 24.0±4.6 and 24.19±4 years, respectively (P=0.88), mean scores of husband's support in the intervention and control groups were 8.23±1.6 and 8.06±2.6, respectively (P=0.46), and mean scores of others' support were 8.26±2.2 and 8.09±2.4 (P=0.93). Chi-squared and Fisher's exact tests showed that the two groups were similar in terms of maternal educational level (P=0.34), socio-economic class (P=0.70), occupation (P=0.37), history of abortion (P=0.70), status of marital satisfaction (P=0.29), and housing status (P=0.93) (Table 1).

Considering the results of independent t-test, mean score of maternal-fetal attachment was similar in the two groups before training (P=0.966), while it was significantly different post-intervention (P=0.046). Based on paired t-test results, in the intervention group, there was a significant difference in mean scores of maternal-fetal attachment before and after training (P<0.001), the difference was also significant in the control group (P=0.005). Independent t-test reflected that mean score of changes in maternal-fetal attachment was significantly higher in the intervention group than the controls (P=0.004; Table 2). Moreover, using covariance analysis and control of maternal-fetal attachment before the intervention, significant differences were observed between the two groups in terms of the effect of training on maternal-fetal attachment (P<0.001, t=16.14, B=94.61).

According to independent t-test results, mean depression score before training was similar in the two groups (P=0.744), but it was significantly different after the intervention (P=0.004). Based

 on paired t-test, in the intervention group, there was no statistically significant difference between the mean depression scores pre- and post-intervention (P=0.151); however, the difference was statistically significant in the control group (P<0.001; Table 3)

According to independent t-test, mean scores of stress before training were similar in the two groups (P=0.300), but the difference was statistically significant after training (P=0.002). Based on paired t-test results, in the intervention group, statistically significant differences were noted between stress scores before and after the intervention (P<0.001), but the difference was not significant in the control group (P=0.88; Table 4).

Table 5. Mean anxiety score before and after training during pregnancy in the intervention and control groups

According to independent t-test results, mean anxiety score was similar in the two groups before training (P=0.143), but the difference was statistically significant after the intervention (P<0.001). Additionally, paired t-test showed a significant difference in mean anxiety score before and after training in the intervention group (P<0.001). The difference was statistically significant in the control group, as well (P<0.001; Table 5).