This research is an interventional study of a randomized controlled trial with two parallel groups (Intervention and control groups).
Participants
This study was conducted in Shahid Akbarabadi Educational and Medical Center in Tehran, Iran. The criteria for entering the study included the reading and writing literacy, being married, being Iranian, no medical prohibition for pregnancy, no children, desirable lost pregnancy, no self-reported of taking drugs and psychotropic substances, not having any mental illnesses at present or in the past as expressed by the participant, no pregnancy after the failed pregnancy, pregnancy loss history of up to 2 cases, and not having infertility problems. The criteria for being excluded from the study involved occurrence of any stressful events (divorce, financial crisis, death of the immediate family members, etc.) during the study, not attending at least two sessions of training sessions, and becoming pregnant during the study.
Study Instrument
Data collection tools included demographic questionnaire, containing 17 questions (age, education and employment status of couples, economic status, duration of the marriage, number of pregnancies, age of women during the first pregnancy, cause of failed pregnancy, the gender of the lost fetus, number of lost pregnancies, delivery type, and the current preferred method of contraception) and O'Connor's Decisional Conflict Scale, which includes a question about the person's choice of pregnancy in the future, and 16 questions with five decisional sub-domains (three questions about the informed subscale, three questions about value clarity, three questions about decision support, and three questions in the uncertainty sub-domains, and finally four questions about an effective decision. According to the questionnaire guide, the questions were answered with a 5-point scale yes (0), possibly yes (1), uncertain (2), possibly no (3), and no (4). In each of the sub-domains, the sum of the scores of the questions was first calculated and then divided by the number of questions and after that, the division result was multiplied by 25, and the range of scores was from zero to 100. In summing up the scores, the closer the total score was to 100, the more the conflict signs (uncertain and delayed decisions) were, and the lower the score was, the less the decisional conflict (the more confident decisions) was (zero was a sign of non-conflict in decision-making and 100 was a sign of maximum decisional conflict).
The original version of the decisional conflict questionnaire was developed by O'Connor (1993) involving 16 questions to review decision-making, and was revised in 2005 and updated in 2010. For the original version of this tool, optimal validity and correlation were reported with the optimal re-testing method and the Cronbach's alpha coefficient was 0.78 [22]. In a study by Ghiyasvandian et al. (2013) the validity of the decisional conflict scale was calculated with Cronbach's alpha coefficient of 0.94 [23]. Moudi et al. (2018) reported tool reliability as 0.92 [24].
In this study, the content validity was used to determine tool validity. For face validity, the decisional conflict tool was given to 20 women with a failed pregnancy and the levels of difficulty (difficulty in understanding words and phrases) and ambiguity (possible misunderstandings of phrases or semantic incompatibility of words) were identified. The reliability of this study was evaluated by two methods of Cronbach's alpha and test re-test and tools were given to 20 women with failed pregnancies, and then were given to them again two weeks later and the internal correlation of the questions was calculated with Cronbach's alpha coefficient of 0.869. It was 0.749 for the subscale of awareness, 0.734 for the subscale of individual values, 0.655 for decision support, 0.710 for confidence, and 0.778 for effective decision-making. Repetition by Pearson correlation coefficient for the total instrument of decisional conflict was 0.986 and 0.946 for subscales of awareness, 0.967 for recognition of individual values, 0.901 for decision support, 0.946 for confidence, and 0.993for effective decisions.
Interventions
This study was conducted from January 2020 to May 2020 on 80 women who had been hospitalized during 3-6 weeks ago due to fail pregnancy. To select the samples, with the permission of the hospital management, the researcher continuously referred to the hospital's medical records unit, extracted the names of women with failed pregnancies in the last 3-6 weeks. Then called them by phone and, selected the eligible participants to enter the study and after explaining the objectives of the study, and the educational content and the number of training sessions and the possible results and applications for the participants, invited them to participate in the project on a specific day, date and time. The participants in the project were assured that all information was confidential and the written informed and free consent was obtained from them. Prior to the start of the study, the demographic questionnaire and the decisional conflict scale were distributed to be complete by all participants (control and intervention) in an appropriate position, and a quiet place in the research environment. In cases where the research units needed explanations for the completion of the questionnaire, the guides were provided by the researcher. After collecting the questionnaires before the intervention, the implementation method of the plan and its purpose were described in details to the samples. The educational content was prepared using reference books and related descriptive articles and its validity was approved by 5 faculty members of Iran University of Medical Sciences. Samples of the control group were asked to leave the hospital and wait for a re-invitation in the next 6 weeks to complete the questionnaires again. On the same day, the first empowerment training session was performed based on the four steps (threat perception, problem solving, educational participation, evaluation) of the four 90-minute sessions (training sessions, and group discussion) over 4 consecutive weeks in 4 groups of 8-12 people in the Akbarabadi educational and medical center in a training class. The date and time of the next meeting were set with the intervention group. In this way, the empowerment support training program was implemented in groups through questions and answers, slide shows, and presentation of pamphlets to be studied by husbands at the end of each session. To increase the threat perception of the samples, it was necessary to increase their level of knowledge and understanding of a failed pregnancy and its reasons, which was improved during the training sessions. Problem-solving sessions were conducted with the participation of a team of midwifery and psychology specialists, and they were assisted by providing coping strategies, adaptations, and introductions of selected options, as well as finding solutions by the women themselves and increasing their sense of self-efficacy. Knowledge and ability and support helped them to increase their self-esteem. With increasing self-efficacy, self-esteem improved self-control. To increase educational participation, at the end of each session, an educational pamphlet was provided to the participants to transfer the content to the family, especially spouses. To implement the fourth step of the empowerment model (evaluation), the researcher evaluated the process during the sessions and at the beginning of each session (examining the degree of listening, attention, learning, interest, and mental participation) and ultimately the final evaluation was performed 6 weeks after the end of the intervention by re-completing the decisional conflict scale by both control and intervention groups. (Table 1)
Sample Size
The sample size was calculated using the following formula:
The sample size was calculated to involve 32 women, taking into account the level of confidence of 95% and the test power of 80% and considering Cohen's d the effect size on the average on the decisional conflict variable in the intervention group compared to the control group (ES = 0.7); it was finalized to involve 40 people, including a 25% drop in the sample size in each group.
Randomization
Each participant was given a number according to the list of eligible participants in the study, and provided a random list for two groups from Sealed Envelope site [25]. In the sealed envelope site we used simple randomization service base on block identifier, block size, sequence within the block, and group code. A list created and generate a unique randomization code. With block sizes: 4, 6, 8. we used a saved copy of this list to look up the intervention group for the given code. Eligible women with failed pregnancies based on a random list, assigned to the control (routine group, n=40) and intervention (empowerment group, n=40) groups. On the day of the clients' visit, the list was given to a colleague non-involved in the plan (secretary of the clinic) to put the names of the samples in order of entering the meeting place in the list. In this way, the samples were placed in two groups of intervention and control completely by accident and without knowing about that classification, neither by the participants nor by the researcher. The researcher’s full name and telephone number was provided to the subjects to inform the researcher in case of a problem or any question.
Statistical methods
Data analysis was performed using SPSS version 16 and the descriptive statistics such as frequency tables, numerical indicators and inferential statistics such as Fisher's statistical tests, chi squared, independent T-test, and paired T-test were performed, and the significant level in this study was considered as (p < 0.05).