Type of study:
This is a semi-experimental and interventional study conducted on two intervention groups and one control group. In this study, the maternal and neonatal outcomes of the care provided by a doula and a trained companion were compared with the usual midwifery care in the labor and birth unit. In this article, only the maternal outcomes are discussed and the neonatal outcomes will be discussed in another article.
Participants:
The study population included 150 pregnant women who attended the clinics of public and private hospitals of Arak, which is one of the largest cities in central Iran and is the capital of Markazi province. The area of this city is about 98.7178 km2 and covers 42.4% of the total area of the province. The population of this city is 591,737 according to the 2016 national census. This city has public maternity hospitals including Imam Khomeini, Taleghani and Ebne-Sina hospitals, which offer doula services if requested by the mother.
Inclusion and exclusion criteria:
Criteria for entering the study included: low-risk pregnant women aged 18 to 40 years, gestational age of 38 to 42 weeks based on accurate LMP or ultrasound, spontaneous onset of labor pains, and healthy, live, single fetus with display of head and normal cervical delivery conditions according to midwife or physician. The criteria for not entering the study included: any known physical disease related or unrelated to pregnancy, mental illness of mother, known fetal abnormalities, previous history of cesarean section or desire to have selective cesarean section, and history of giving birth to infant who weighs less than 2500 or more than 4000 grams. The exclusion criteria included a reluctance to continue with the study.
Sample size:
The sample size was determined to be 45 mothers in each group according to the studies of Khavandizadeh Aghdam (19), and Samieizadeh Tusi(20) as well as Nelson’s neonatal book(21), considering 95% confidence and 80% test power. Then, taking into account the possibility of sample’s drop and using the sample size formula, it was decided to add 5 samples to each group, so a total of 150 mothers were examined in this study.
Method:
The trained (lay) companion group contained 50 pregnant women with 35-37 gestational weeks confirmed by accurate LMP or ultrasound from 10 pre-natal clinics of Arak public Hospitals including Arastoo, Imam Khomeini, Jahangiri, Zeinab Kobra, 30-metri Miqan, Gavkhaneh, Hejrat, Shohadaie Safari, Valiasr, and Karahrud. The centers were selected from different parts of the city with clients coming from all economic and cultural classes who receive free services. To select the samples, we entered the SIB system in the mentioned centers and extracted the list of pregnant mothers who had the inclusion criteria from January to March 2019. Then the mothers were asked to enter the study and those who agreed were randomly selected. We used simple random sampling for choosing mothers from these centers. We had a list of all mothers of this center (from 1 to 389), then a number for each mother was randomly generated by using Excel, and the 50 samples that we required (from number 1 to 50) were then taken. After this, the necessary information about the objectives and method of study were given to them. The mothers were asked to nominate a person of their choice to accompany them during the labor and birth. A trained person should have at least one experience of childbirth. After being selected by the mother and introduced to the research team, they were invited to participate in educational classes, during which they were taught how to care for the mothers during labor. They were also assured that they would be notified of the time of the classes. After sampling was completed in this group, two training sessions were designed and the time of these sessions was coordinated with the mothers and their companions. Since a number of samples were excluded from the study for reasons such as not attending the training sessions, same number of eligible people were randomly assigned to the study to prevent any problem in statistical analysis. The mothers of this group finally gave birth in Imam Khomeini and Taleghani hospitals, according to their choice (Sina Hospital is private and Imam Khomeini and Taleghani Hospitals are public).
In the sessions, topics such as stages of labor, symptoms of labor, measures needed to facilitate labor, and role of the companion before coming to the hospital including: helping a pregnant woman to go to the hospital if she sees any signs of labor, continuous presence beside the patient, gaining the trust, building a sincere relationship, listening to the mother's concerns and worries, encouraging proper breathing techniques, using massage or applying cold or hot compresses (according to the mother's wishes), teaching relaxation methods, giving simple and understandable information about the birth process, encouraging frequent emptying of the bladder, helping and encouraging mobility, changing the position during the labor and using the delivery ball if possible, keeping the mother informed of the progress and birth process, helping the mother to get comfortable, encouraging the mother to push when needed, boosting her mood, preparing the mother for childbirth, advising the mother to work with the birth agent and teaching her how to breathe properly, applying pressure when infant is coming out, giving information about the importance of skin-to-skin contact with the baby immediately after the birth and the importance of early breastfeeding, were taught.
At the last session, the mothers and their companions were asked to inform the researcher when they see signs of childbirth, so that coordination could be made with the delivery and obstetric unit of Imam Khomeini and Taleghani Hospitals to accept them. If a woman with gestational age of under 37 weeks entered the delivery phase, she was excluded from the study, although we did not encounter such a sample. Also, if a woman had dilatation of 3 cm or more (active phase), she was admitted to the hospital.
Doula's group contained 50 pregnant women aged 35 to 37 years, who had been selected from the counseling centers of Madaryar and Mehr-Madari using convenience sampling. According to the instructions of the Ministry of Health, the cost of doula in these centers is 50,000 Tomans ($ 4.2) per hour. Most of the clients who receive doula services are nulliparous women, and women who deliver their second child are more likely to apply for these services due to their difficult previous labor. In these centers, a doula is considered for all mothers. Doula is one of the center's staff who personally takes all responsibilities of a doula during labor, but the delivery is done by other staff member.
62 mothers who had been registered for doula were selected from three Taleghani, Sina and Imam Khomeini hospitals from July 2018 to June 2019. Like companion group, simple random sampling was used for choosing mothers for this group. After providing a list of all mothers of this center (from 1 to 62), we randomly generated a number for each mother by using Excel and then 50 samples were taken (number 1 to 50). Finally, our samples gave birth in Imam Khomeini and Sina hospitals, and this was their choice, and no restrictions were considered.
According to the coordination made with the doulas, they were supposed to accept their patients in the active phase in the hospital, inform the researcher, accept the admission of each patient in the center, and teach mothers according to their own routine during the labor.
Samples of the control group were from mothers who attended the Taleghani Hospital for prenatal care. Like doula and companion group, we used simple random sampling for choosing mothers for control group, from September 2018 to October 2018. Accordingly, a list of all mothers of this hospital who had our criteria (from 1 to 95) was provided then we randomly generated a number for each mother by using Excel and 50 samples (number 1 to 50) were then taken from 95 mothers. Selected mothers were asked to inform the researcher to coordinate with the hospital as soon as the labor pains began. When the women were admitted to the hospital, they were cared for by midwifery staff as usual. In this case, 5-6 midwives care for 6 maternity beds in each shift, and midwives are not required to go to see a pregnant woman in her bed when they have nothing to do with her delivery. Since Taleghani Hospital is a teaching and referral hospital, samples of the control group were shared jointly by midwives and assistants. In both trained companion and doula groups, supportive care began at the time of admission to the hospital and continued uninterruptedly until delivery and up to one hour after delivery. We could not add an explanation to the summary due to word limitation
Data collection tools:
Data collection tools included:
1. A researcher-made checklist containing questions related to demographic information, fertility history and childbirth outcomes including the duration of active phase of the first stage (from 3 to 10 cm dilatation) and the duration of second stage of delivery (from 10 cm dilatation to complete exit of placenta), which was completed by the researcher. To determine the scientific validity of the first part of the tool (information checklist), the content validity method was used. For this purpose, 10 faculty members of the School of Nursing & Midwifery of Tehran University of Medical Sciences were consulted and according to their comments and suggestions, the necessary corrections were made and the final version of the tool was prepared and used with the approval of the supervisor. Simultaneous assessment method was also used to examine the reliability of information registration form. For this purpose, the researcher recorded the results of clinical examinations of several patients and then a midwife was asked to do the same examinations. After this, the researcher compared both results and if the similarity between them was above 80%, the scientific reliability of information registration was confirmed.
2. The Spielberger State-Trait Anxiety Inventory (STAI). In this tool the score of 20 to 31 indicates mild anxiety, 32 to 53 shows moderate anxiety, 64 and higher indicates relatively severe anxiety, and score of 76 and above indicates very severe anxiety. The validity and reliability of this questionnaire have been confirmed in 1994 by Mahram who studied 600 people(22). The questionnaire was completed in two dilatation times of 3-4 cm and 8-10 cm by the researcher through an interview with the mother, and then the difference between the mother's anxiety score in both times was calculated.
3. Visual Analogue Scale (VAS) for pain which uses a 10 cm calibrated ruler. The score of ten indicates the most severe pain and score of zero indicates no pain. The amount of pain is determined by the patient using the ruler. In several studies, the scientific validity and reliability of this tool have been confirmed. (23, 24).In Iran the reliability of this scale with the correlation coefficient of r = 0.88 has been confirmed. The pain scale was presented to the mothers at the time of admission and every hour until the time of delivery, and they were asked to indicate the severity of their pain by selecting one of the numbers.
4 - Maternal satisfaction rate was assessed 24 hours after the delivery with the revised satisfaction birth questionnaire by Hollins Martin (BSS-R). The revised birth satisfaction scale (BSS-R) was designed in 2011 by Caroline Hollins Martin et al. This revised questionnaire includes 10 questions in 3 areas of mother's satisfaction with the quality of care provided, her personality traits, and the anxiety experienced during labor and childbirth(25). In Iran, Rahimi Kian et al. (2017) have validated this questionnaire(26). To express their satisfaction, the samples responded by choosing the options: I disagree, I have no opinion, I agree and I agree very much. For these responses, a score of 0 to 4 was considered, respectively, based on the 5-point Likert’s scale
Data analysis:
The data were analyzed after collection by SPSS statistical software (22) using Kruskal Wallis, Chi-Square, ANOVA and Fisher’s exact tests. ANOVA method was used to moderate the effect of factors whose distribution was not homogeneous in the three study groups.
Ethical considerations:
The proposal of this project was approved by the Ethics Council of Arak University of Medical Sciences with the code of ethics: IR.TUMS.FNM.REC.1397.097. It was also registered in the Iranian Clinical Trials Registration Center.
Findings:
Study participants in the three groups of doula, trained companion and control, were in the age range of 23 to 30 years (60%, 58% and 52%, respectively). The majority of mothers in the three groups had diploma (82%, 64% and 42%, respectively). Most of the mothers were nulliparous (78%, 64% and 52%, respectively) and their infant weight was mainly between 2,500 and 4,000 grams (92%, 98% and 96%, respectively). The demographic and midwifery characteristics of all samples are presented in Table 1. The companion chosen by the mothers in the trained companion group was one of the family members, including the mother or sister of the pregnant woman.
Table 1. Demographic characteristics and obstetric history of participants
|
p-value
|
Model of care in intervention and control groups
|
Variable
|
Doula
|
Trained companion
(family member)
|
Maternity ward midwife
n (%)
|
0.222
|
5(10)
30(60)
15(30)
|
12(24)
29(58)
9(18)
|
13(26)
26(52)
11(22)
|
18-22
23-30
31-38
|
Age (year)
|
0.001>
|
1(2)
7(14)
41(82)
1(2)
|
5(10)
13(26)
32(64)
0(0)
|
10(20)
19(38)
21(42)
0(0)
|
Illiterate & elementary/secondary school
Diploma Bachelor's degree and higher
|
Educational level
|
0.001>
|
44(88)
6(12)
|
45(90)
5(10)
|
24(48)
26(52)
|
City
Village
|
Place of residence
|
0.054
|
39(78)
10(20)
1(2)
|
32(64)
13(26))
5(10)
|
26(52)
15(30)
9(18)
|
First
Second
Third
|
Parity
|
0.687
|
1(2)
46(92)
3(6)
|
0(0)
49(98)
1(2)
|
1(2)
48(96)
1(2)
|
2500>
250-4000
4000&more
|
Infant’s birth weight(gr)
|
The average duration of active phase of labor was 279±94.37 min in the control group, 234.68±118.74 min in the doula group, and 256.66±108.75 min in the lay companion group. The results of Kruskal Wallis test showed no statistically significant difference between the three groups in this regard (p-value = 0.063). However, when the relationship between the groups and the duration of active phase of labor was evaluated by ANOVA test and also confounding factors were controlled, this relationship was significant (p-value = 0.22).
The mean duration of second stage of labor was 22.75 ± 30.57 min in the control group, 10 ± 5.61 min in doula group and 10.35 ± 5.1 in the lay companion group. The difference between these variables was statistically significant. Therefore, the duration of second stage of labor was completely affected by the type of group (p-value<0.001), (Table 2).
Table 2. Comparison of duration of first and second stage of labor in three study groups
p-value
|
Doula
|
Trained companion
|
Maternity ward midwife
|
Variable
|
0.063
|
234.68±118.74
|
256.66±108.75
|
279±94.37
|
Active phase duration(minute)
|
0.001>
|
10±5.61
|
10.35±5.1
|
22.75±30.57
|
Second stage duration(minute)
|
Kruskal Wallis * Four-way ANOVA*
In order to measure the severity of pain, considering that most samples (101 people) gave birth during the first 5 hours of intervention (the severity of pain was not calculated for those who had cesarean section), the test could not be carried out between the three groups after the fifth hour, so the pain was measured at admission and then 1, 2, 3, 4, and 5 hours after the admission. Comparing the severity of labor pain in the three study groups showed that, the mean pain score at the time of admission was 60.21 in the control group, 71.14 in the doula group and 62.69 in the trained companion group based on the VAS scale and the difference between these scores was not statistically significant. The mean pain score in the first hour after the intervention was 67.80 in the control group, 67.51 in the doula group and 59.52 in the trained companion group, and the difference between these scores was not statistically significant. The difference between mean pain scores in the second, third, and fourth hours was also not statistically significant. It means that, the severity of pain in the fourth hour of labor was not affected by the type of group and the care model (Table 3). However, the mean pain score in the fifth hour after the intervention in the control group was 33.85, in the doula group it was 20.93 and in the trained companion group it was 19.81, and the difference between these scores was statistically significant. Also, when four-variate ANOVA test was used to evaluate the relationship between the groups and severity of pain at the fifth hour and also the confounding factors (place of residence, education, and number of pregnancies) were controlled, statistically significant difference was not obtained for this relationship (p-value = 0.242).
Table 3. Comparison of the pain severity in the three study groups
p-value
|
Doula
|
Trained companion
|
Maternity ward midwife
|
Variable
|
0.337
|
71.14
|
62.69
|
60.21
|
Severity of pain at admission
|
0.505
|
67.51
|
59.52
|
67.80
|
Severity of pain in the first hour of labor
|
0.409
|
59.78
|
57.49
|
67.51
|
Severity of pain in the second hour of labor
|
0.07
|
47.60
|
48.96
|
62.40
|
Severity of pain in the third hour of labor
|
0.176
|
33.68
|
36.91
|
44.29
|
Severity of pain in the fourth hour of labor
|
0.003
|
20.93
|
19.81
|
33.85
|
Severity of pain in the fifth hour of labor
|
Kruskal Wallis * Four-way ANOVA*
Comparison of the difference between anxiety score of mothers in two stages of labor in the three groups showed that, the mean difference in anxiety score in two stages of labor in the control group was 33.53, in the doula group it was 74.51 and in the trained companion group it was 80.08 and also the difference in these scores was statistically significant (p <0.001) (Table 4). Therefore, the level of anxiety during labor was completely influenced by the type of group and the care model.
Table 4. Comparison of the mean anxiety score in two stages of labor in the three study groups
p-value
|
Doula
|
Trained companion
|
Maternity ward midwife
|
Variable
|
<0.001
|
74.51
|
80.08
|
33.53
|
Manifest anxiety of mothers
|
Kruskal Wallis*
To compare the satisfactory distribution of childbirth in the three groups of control, doula and trained companion, Kruskal Wallis test was used, the results of which are shown in Table 5. The mean score was 51.94 in the control group, 89.11 in the doula group and 85.45 in the trained companion group. The difference in these scores was statistically significant (p <0.001) (Table 5). Therefore, the condition of maternity satisfaction was completely affected by the type of group and the care model.
Table 5. Comparison of satisfaction with delivery in the three study groups
p-value
|
Doula
|
Trained companion
|
Maternity ward midwife
|
Variable
|
<0.001
|
89.11
|
85.45
|
51.94
|
Satisfaction
|
Kruskal Wallis*