Study design and participants
This quasi-experimental study was conducted with a pretest-posttest design and a control group. Participants were ninety mothers of premature newborns selected from the NICUs of Zeinabieh and Hafez hospitals affiliated to Shiraz University of Medical Sciences, Shiraz, Iran. Inclusion criteria were ability to speak, read, and write in Persian, no history of newborn’s hospitalization in NICU, agreement for participation, and a premature newborn with a gestational age of 26–36 weeks, a birth weight less than 2500 grams, and no life-threatening problem. Exclusion criteria were newborn’s death, discharge from NICU, or transfer to another healthcare center during the study, voluntary withdrawal from the study, and any absence from the intervention sessions. Participants were randomly allocated to a control and an intervention group using a computer-generated randomization list. Accordingly, ninety sealed envelopes each containing a paper with a number from 1 to 90 were prepared. Numbers had randomly been generated using a computer program. Each participant randomly selected an envelope and reported its number to a statistician who knew the group of each number. As the intervention of the study was a training program, blinding was not possible.
Sample size was calculated using the results of a former study which reported that the mean score of anxiety in mothers was 91.4±3.16 in the control group and 61.2±4.4 in the intervention group[10]. Accordingly, with a confidence level of 0.95, a power of 0.80, a minimum mean difference of 2.83, and a probable attrition rate of 10%, the sample size calculation formula (Figure 1) showed that 45 participants per group were needed.
Measurement
Data were collected using a demographic questionnaire and the Parental Stress Scale. The items of the demographic questionnaire were on participants’ age, occupation, educational level, marital status, number of children, type of delivery, and their newborn’s gestational age, gender, and weight. Demographic data were retrieved from participants and the medical records of their newborns. The Parental Stress Scale has 26 items in three subscales, namely NICU environment (five items), the appearance and behaviors of the newborn (fourteen items), and parent-newborn relationships (seven items). Items are scored on a five-point Likert scale as follows: 0: “Not at all stressful”; 1: “A little stressful”; 2: “Moderately stressful”; 3: “Very stressful”; and 4: “Extremely stressful”. Therefore, the possible total score of the scale is 0–104, with higher scores showing greater stress. Scores 0–20, 21–40, 41–60, 61–80, and 81–104 respectively show very low, low, moderate, high, and extremely high stress. This scale has acceptable validity and reliability with a Cronbach’s alpha of 0.86[11]. Study participants completed this scale before and one week after the study intervention.
Intervention
Participants in the control group did not receive any EI-related intervention, while their counterparts in the intervention group participated in six 1.5-hour EI skills training sessions held twice weekly. A psychologist provided EI skills training to participants in a room next to NICU in the study setting while the first author supervised the sessions. The training program was developed based on Mayer and Salvi’s EI model [12] and its content was as follows:
Session 1: After an introduction, participants were informed about the intervention and the number and duration of the sessions, and then, their questions were answered. Finally, diaphragmatic breathing and mental imagery were trained and exercised.
Session 2: EI and its difference with intelligent quotient were discussed.
Session 3: Participants were provided with explanations about self-awareness, discovery of emotions, different simple and complex emotions and their differences, and how to understand the emotions of self and others.
Session 4: Explanations were provided about understanding of emotions, empathy, social skills, and establishment of effective communication.
Session 5: This session was on the regulation and management of emotions.
Session 6: Verbal and written summaries of all trainings in previous sessions were provided and participants’ questions were answered.
A panel of experts consisted of two psychologists and four psychiatric nurses approved the protocol and the content of the training program. No change was made to the approved program during the study.
Data analysis
Data were analyzed using the SPSS software (v. 16.0, SPSS Inc., Chicago). Frequency, mean, and standard deviation were calculated to describe the data. The Kolmogorov-Smirnov and the Shapiro-Wilk tests were performed for normality testing. The independent-sample t, the Chi-square, and the Fisher’s exact tests were used for between-group comparisons, while the paired-sample t and the repeated measures analysis of variance were used for within-group comparisons. The analysis of covariance was also used for between-group comparison respecting the posttest mean score of stress adjusted for the effects of potential confounders. The level of significance was set at less than 0.05.
Findings
In total, ninety mothers of premature newborns in NICU were recruited to the study (Figure 2). Eight participants from each group were excluded due to their lost to follow-up and the study was completed with 37 participants in each group. Study groups did not significantly differ from each other respecting participants’ and their newborns’ characteristics (P > 0.05), except for newborns’ birth weight and the existence of premature newborns in the family (P > 0.05) (Table 1).
There was no significant difference between the study groups in terms of the pretest mean score of stress (P > 0.37), while the posttest mean score of stress in the intervention group was significantly less than the control group (P < 0.001). The results of the repeated measures analysis of variance also showed the significant effects of time, group, and time-group interaction (P < 0.001; Table 2).
The analysis of covariance was used for between-group comparison in terms of the posttest mean score of stress adjusted for the confounding effects of newborns’ birth weight and the existence of premature newborns in the family. Its results showed that after removing the effects of these potential confounders, the posttest mean score of stress in the intervention group was significantly less than the control group (P < 0.001) (Table 3).