This Quasi-experimental study was conducted in the NICU of Mofid children’s medical centre, Tehran, Iran from November 2018 to April 2019.

The Shahid Beheshti University of Medical Sciences Research Ethics Committee (IR.SBMU.RAM.REC.1394.352) granted ethical approval and the Iranian Registry of Clinical Trial code is IRCT 20190617043912N1. Approximately 129 new newborns are admitted to NICU each year and 30% of them are premature. For the sample size according to previous study, with a confidence level of 95% and test power of 90%, 60 individuals were selected for this research, and they were divided into 30 participants for each group [16].

The randomisation was performed by the central randomisation unit at the Neonatal Health Research Centre, using block randomisation with the same block sizes. All participants were allocated a number. Then the randomisation numbers were given to the head of nursing staff. Thus, she allocated participants to intervention and control groups randomly. All participants provided written informed consent before the data collection; this included permission to use in publications.

All parents of infants admitted to neonatal intensive care units, have access to the telephone and no hearing or speech problems.

Parents who showed unwillingness to participate in the study, not able to understand and speak Persian and some who had no access to telephone services.

First, after admission into the NICU, the assistant researcher visited the parents and explained the research objectives. Then an informed consent was attained from the parents in both intervention and control groups, and the demographics form was completed for the parents and neonates.

In 15 consecutive days, the researcher called to the parents (each day, one call) and announced information to family member about recent vital signs of the neonate’s respiration, consciousness level, sleeping rhythm, feeding, and the priority of physical attendance of his/her parents or absence of this necessity. The Beck Anxiety Inventory previously available to family members was filled out one time before intervention, and subsequently after phone contacts in first, fifth, tenth and fifteenth day of their neonates’ admission by self-reporting method. Control group did not received telenursing intervention. Finally, the anxiety level of the participating parents in both the groups was assessed by using the beck anxiety inventory [Table/Fig-1].

[Table/Fig-1]:

Study plan.

Content validity of beck anxiety inventory was assessed and confirmed by 10 nursing academic board members, Shahid Beheshti University of Medical Sciences. For assess reliability, two methods were used: internal reliability and external reliability. To consider internal reliability, α-Cronbach’s index was used and computed 96% in the present study and to consider external reliability identified as re-test method, it was used Interclass Correlation Coefficient (ICC) that obtained 85%.

An informed consent was signed and demographic questionnaire was filled out by the parents, 24 hours after hospitalisation of their neonate in NICU. In order to respect the confidentiality of the participants’ name, phone number of one of parents was recorded. This phone number is in access to researcher and saved as a special code with the recommended name of that member. Researcher has called only by one phone number that previously announced to family member.

All of phone calls were done during 16-16:15 in the evening. At the beginning of each conversation, the parents were asked to repeat parts of the former conversation to ensure that the proper understanding of previous conversation was attained. The average duration time of the conversations was 15 minutes.

The telephone intervention procedure was standardised and also recorded for each parent based on Cox’s Interaction Model of Client Health Behaviour (IMCHB) [17]. It involved a systemic comprehensive assessment of the participant’s requirements, including basic features like the type of participant’s characteristics, the date, time, and nature of the call, as well as a detailed description of the symptoms, problems, or the call reasons. Moreover, all of the present study premature neonates’ required NICU stay for 15 days as recommended by the physicians. Eight of premature neonates’ require NICU more than 15 days. At the end of the intervention, a farewell ceremony was held for the parents and their neonates.

The samples were described by using descriptive statistical methods including the mean and the Standard Deviation (SD). The descriptive statistical method’s findings were calculated based on both qualitative and quantitative data. Shapiro-Wilk test was used in order to determine the normal distribution of the data. Taking into account, the homogeneity of the qualitative factors in two groups, the mean score of the variables were compared between the two groups. This was done by using a parametric t-test, and also Wilcoxon test for non-parametric. Data analyses were performed in SPSS version 21.0 software.

Sixty neonates’ parents were enrolled in this Quasi-experimental study. Minimum, maximum, and average ages of parents were calculated as 20, 39, and 29 years, respectively. Most of the participants were mothers (83.3%), and the majority of them had an education of a high school diploma (43.3%). The majority of neonates were girls, at 83.3% [Table/Fig-2].

The [Table/Fig-3] shows the average anxiety scores at different stages of the research. Based on this table, at the beginning of the study (the first day of admission), there was no significant difference between the anxiety scores of the control and intervention groups (p>0.05). In other hands, The comparison mean of anxiety scores in the control and intervention groups in the first, fifth, tenth and fifteenth day of admission showed a significant difference at the level of one percent error between two groups (p<0.001). Also, the p-value for the comparison of the mean total score during the 15 days of admission, which is equal to -8.44, is significant at a 1% error level.

According to the results of the Wilcoxon test in [Table/Fig-3], showed that daily conversations effectively reduced anxiety at 1% error level (p≤0.001) [Table/Fig-4].

The results of comparing the mean scores of parents’ anxiety levels between the intervention and control groups indicated that telenursing made a significant statistical difference in the intervention group.

Findings of a study showed that the effectiveness of active telephone support on breastfeeding premature infants in Sweden in 2013. They concluded that telenursing improved breastfeeding, maternal satisfaction, mother and infant attachment, and also alleviated maternal anxiety [18]. In another study, the effect of phone intervention by nurses on the mood and anxiety level of patients with implantable cardioverter defibrillators and their caregivers was assessed. They stated that eight weeks of telephone follow-up decreased the anxiety level considerably [19]. Also, a study was concluded that telenursing had a positive-effect on increasing attachment and mitigating the level of stress in mothers with premature infants [7].

Results of a study established that telenursing via the Internet was more effective and has fewer costs than the conventional hospital-based follow-up [17]. A study showed that telemedicine reduces care costs and enhances emotional support during and after hospitalisation [20].

Another study described a significant statistical relationship between informing families about patients’ conditions and reducing their anxiety level after phone intervention [13]. Telenursing can be effective as a treatment plan for accelerating recovery after coronary artery bypass grafting [21].

In the present study, the mean score of primary anxiety before phone calls in the intervention group and in the control group was 51/80±11/2 and 52.45±11.9, respectively. However, after intervention, score of anxiety improved in intervention group. This finding of the present study is to coordinate with above studies. It seems that resolving information’ gap in patient’s families by use of telenursing can improve their anxiety level.

Use of Beck Anxiety Inventory is a limitation due to its probable bias during self-reporting. Also, the different cultural context of participants in phone conversations can be effective in intervention output.

The telenursing intervention was effective in reducing anxiety in parents who had their premature neonates in NICU for 15 days. Given the easy access to a telephone for almost all parents and its simple use, including this technology in the educational and care context of neonatal parents is recommended. Also, using this technology requires establishing remote communication methods courses in educational programs of nurses. In its context, this research provides the ground for further studies and technology introduction to the nursing field. In addition to that, the technology can be used in planning in order to cut and minimise the costs, caused by temporal and spatial limitations.