1. Background
Breast milk is considered the best food for babies. Exclusive breastfeeding (EBF) for six months can decrease by 13 % of children's mortalities in developing countries. Non-exclusive breastfeeding may cause 55% and 53% of diarrheal and acute respiratory deaths. In fact, 1.24 million deaths that occur during the first 6 months of life are associated with non-exclusive breastfeeding (1-7). Universally, the results have shown that only 35% of babies were exclusively breastfed during their first 4 months of life (7-9).
In 1997, the National Breastfeeding Center merged to “Children’s Health Office”. According to the Iranian demographic and health survey, the practice of mothers in breastfeeding improved (10). Later, the Iranian multiple indicator demographic and health survey showed that EBF increased to 27.8% and 62.8 % in the urban and rural areas, respectively (10).
Considering the significant difference between WHO recommendations for exclusive breastfeeding and its practices in Iran and on the world, the educational interventions are necessary for promoting the mother's practices (10-15).
The Theory of Planned Behavior (TPB) suggests that individual behaviors are motivated by behavior intentions which is a function of three factors including the individual’s attitude toward behavior, subjective norms, and perceived behavioral control (11-14).
Attitude is a feeling opinion about someone or something that refers to the evaluative effects of the positive or negative feelings of persons in performing a specific behavior. Subjective norms reflect the perceived social pressure from significant others to perform or not to perform a specific behavior (16-18).
Perceived behavioral control refers to the individuals' perceptions of their capability to do breastfeeding. Generally, if individuals have a positive attitude or a strong feeling about reference groups or individuals, such a feeling will approve their breastfeeding, and if they perceive that, they will be able to perform the breastfeeding or have control over their performance (16-18).
In recent years, many efforts have been made to develop the TPB and adapt it for intervening on different behaviors. Legal and illegal drug use, predicting blood donation intentions, predicting the retention of first‐time donors, cancer screening intention, exercise behavior, and predict the intention of EBF were some of the interventions based on the extended theory of planned behavior (19-26).
Moral norm and self-identity were the constructs that Professor Mark Conner added to TBB for predicting the behavior of exclusive breastfeeding. Moral norms can be considered as a person’s perception about the moral accuracy of a behavior. Self-identity may be considered as the relatively continuing specs that people attribute to themselves (16).
2. Objectives
Promoting breastfeeding to reduce the mortality and morbidity in mothers and children demands some innovative programs. Therefore, a theoretical educational program was designed and implemented to evaluate the predictive role of the two constructs of moral norms and self-identity constructs associated with the expanded theory of planned behavior.
3. Methods
In this semi-experimental study, we randomly allocated 20 healthcare houses of 42 primary health care houses into control and intervention groups by a multi-stage sampling method. Then we used two independent samples with continuous outcomes. It was assumed the P < 0.05 as acceptable and a study with 80% power. Therefore, Z1 - α/2 and Z1 - β was gotten the values 1.96 and 0.8416, respectively. The effect size was considered as 80. Calculating for a 10% drop-out rate, one would need to complete approximately 90 patients per arm.
3.1. Inclusion Criteria and Exclusion Criteria
Voluntarily participation and those who were in their third trimester were induced, and those who were not satisfied enough to continue in the research and were absent in the educational program were excluded.
3.2. Data Collection Tool
The data collecting tool was a researcher-made questionnaire containing demographic and specialized questions. The participant’s situations (e.g., pregnancy, age, education, and occupation) were surveyed. Also, specialized questions were developed based on the extended theory of planned behavior and designed according to the exclusive breastfeeding.
The accuracy of the questions was approved by content validity and reliability. The reliability of the questions was surveyed via Cranach’s alpha coefficient test. All of the items have reliability and validity over 70% and 80%, respectively.
To measure the responses to the knowledge questions, for correct, incorrect, and I do not know" answers, 2, 0, and 1 were scored, respectively.
Also, the responses of TPB constructs including, attitude, subjective norm, perceived behavior control, behavioral intention, moral norm and Self-identity were designed based on the guideline of the Theory of Planned Behavior. The questionnaire was designed at seven levels of Likert scales, including extremely agree, quite agree, slightly agree, neither agree nor disagree, slightly disagree, quite disagree and extremely disagree that scored from 1 to 7.
The behavioral question was whether or not the baby was exclusively breastfed in the last six months, assigning 1 and 0 for each yes or no answer, respectively.
3.3. Procedure
After achieving the ethics approval and coordinating the study, the objectives were explained and the written informed consent was obtained from all patients. Then, the research questionnaires were distributed, and the primary data were collected and analyzed.
Based on the need assessment, the educational content, methods, media, and learning sessions were designed and developed. In this study, the main educational method was face to face approach conducted in the health centers. All of the pregnant women were interviewed, and an educational pamphlet was delivered. Based on the educational curriculum for husbands of the women as subjective norms, an educational pamphlet about supporting the role of spouses was designed and distributed.
Also, to promote the mean scores of moral norms and self-identity constructs, According to the social conditions and positive attitude of family members to sharia law and the importance of breastfeeding in the Holy Quran a clergyman and a pediatrician as educators participated in the educational sessions.
For attitude changes, all of the women participated in focus group discussion (5 - 7 women in each group), for perceived behavior control promoting, direct and indirect experiences method were used, respectively. Then in the second educational session, an educational pamphlet was distributed among pregnant women. After delivery, mobile messages associated with exclusive breastfeeding were sent to pregnant women. The data were recollected and compared after six months.
This study was registered in Zahedan University of Medical Sciences with an ethical number of IR. ZAUMS. 1396.56.
3.4. Data Analysis
The Kolmogorov-Smirnov test was used to determine the normality of data; the data analysis was conducted via non-parametric tests.
The mean, percentage, and standard deviation were used to describe the data, while the Mann-Whitney test, Spearman correlation coefficient test, logistic regression, and chi-square test were used to analyze the data. Furthermore, the level of significance for the tests was considered at 0.05.
4. Results
In this study, 168 pregnant women were studied in two groups. The mean age of participants in intervention and control groups were 29 ± 6 and 28.7 ± 5.9, respectively (P = 0.25).
There were no statistically significant differences between the two groups in terms of age, number of pregnancies, live birth numbers and literacy levels (P < 0.05); however, a significant difference was observed between two groups in terms of gestational age, controlled by linear regression (Table 1).
After training, the mean scores of knowledge, attitude, subjective norms, moral norms, self-identity, and perceived behavioral control in the experimental group significantly increased, and a significant difference was observed between control and case group (P < 0.0001) (Table 2).
Table 1.Comparison of the Demographic Characteristics in the Research Groupsa
| Variables | Research Groups | P Value | |
|---|---|---|---|
| Intervention | Control | ||
| Age | 29.2 ± 6 | 28.7 ± 5.9 | 0.61 |
| Number of birth | 2.8 ± 1.3 | 2.5 ± 1.5 | 0.25 |
| Number of live birth | 1.4 ± 0.9 | 1.3 ± 1.23 | 0.72 |
| Employment (women) | 0.96 | ||
| Unemployed | 80 | 78 | |
| Employee | 4 | 5 | |
| Employment (husbands) | 0.38 | ||
| Unemployed | 10 | 12 | |
| Employee | 9 | 14 | |
| Self-employed | 66 | 57 | |
| Education (women) | 0.52 | ||
| Elementary school | 35 | 26 | |
| Secondary school | 15 | 16 | |
| Diploma | 23 | 24 | |
| College education | 12 | 17 | |
| Education (husbands) | 0.491 | ||
| Elementary school | 46 | 38 | |
| Diploma | 38 | 30 | |
| College education | 11 | 15 | |
aValues are expressed as mean ± SD or number.
Based on the results, in demographic characteristics between two groups, no differences were seen.
Table 2.Comparison of the mean scores of the extended theory of planned behavior constructs in the research groups before and after the intervention
| Construct, Research Groups | Intervention | |||||
|---|---|---|---|---|---|---|
| Before | After | |||||
| Mean ± SD | Percent of Total Score | P Mann Whitney | Mean ± SD | Percent of Total Score | P Mann-Whitney | |
| Knowledge | 0.52 | 0.001 | ||||
| Intervention | 12.8 ± 2.2 | 79.8 | 15.3 ± 1.3 | 95.9 | ||
| Control | 12.7 ± 2,64 | 80 | 12.9 ± 2.2 | 80.7 | ||
| Attitude | 0.44 | 0.001 | ||||
| Intervention | 72.2 ± 10.7 | 79.4 | 5 | 85.5 ± 5.2 | 94 | |
| Control | 73.6 ± 8.9 | 80.8 | 73.7 ± 7.8 | 81 | ||
| Subjective norms | 0.18 | 0.001 | ||||
| Intervention | 58.7 ± 8 | 83.9 | 64 ± 5 | 91.5 | ||
| Control | 60.6 ± 6.7 | 86.6 | 60.7 ± 6.2 | 86.7 | ||
| Moral norms | 0.445 | 0.001 | ||||
| Intervention | 30.1 ± 3.6 | 86.1 | 33.3± 2.2 | 95.3 | ||
| Control | 29.5 ± 3.7 | 84.4 | 30.3 ± 3.2 | 86.6 | ||
| Self-identity | 0.493 | 0.001 | ||||
| Intervention | 24.6 ± 2.9 | 88.1 | 27.1 ± 1.1 | 96.8 | ||
| Control | 24.6 ± 2.2 | 87.9 | 24.6 ± 2.1 | 88.2 | ||
| Perceived behavior control | 0.75 | 0.001 | ||||
| Intervention | 28.6 ± 4.3 | 81.8 | 32.5 ± 2.6 | 93.1 | ||
| Control | 28.7 ± 4.5 | 82.2 | 28.8 ± 4 | 82.4 | ||
| Intention | 0.09 | 0.001 | ||||
| Intervention | 30.1 ± 4.5 | 86.2 | 33.2 ± 2 | 95 | ||
| Control | 29.9 ± 3 | 85.6 | 30.9 ± 2.7 | 88.3 | ||
According to Table 2, in the intervention group, the mean scores of all of the constructs compared to the control group after the intervention increased meaningfully, but attitude and self-identity have the greatest (14.6%), and the least (8.7) mean differences, respectively.
Table 3.Comparison of Behavior in Intervention and Control Groups After Educationa
| Group | Exclusive Breastfeeding for 6 Months | No Exclusive Breastfeeding for 6 Months |
|---|---|---|
| Intervention | 75 (88) | 10 (12) |
| Control | 64 (77) | 19 (23) |
| χ2 = 3.64 | P = 0.044 |
aValues are expressed as number (%).
According to Table 3, the study of EBF behavior showed that after childbirth, the rate of EBF in the intervention group was higher than control (N = 11).
Table 4.Logistic Regression Analysis Between Exclusive Breast Feeding and Constructs of the Theory
| Constructs of Model | B | Standard Error | P Value | OR |
|---|---|---|---|---|
| Subjective norms | 0.053 | 0.049 | 0.279 | 1.055 |
| Moral norms | 0.110 | 0.099 | 0.269 | 0.896 |
| Self- identity | 0.039 | 0.149 | 0.796 | 1.039 |
| Perceived behavior control | 0.113 | 0.090 | 0.209 | 1.120 |
| behavioral intention | 0.488 | 0.12 | 0.001 | 1.629 |
According to Table 4, results of the logistic regression analysis between EBF and constructs of the model showed that behavioral intention was the only predictive variable for behavior (P = 0.001), (OR = 1.62).
5. Discussion
In general, the study results showed that theoretical training intervention particularly, the extended theory of planned behavior, increases the EBF as a main and significant behavior in pregnant women which shows the efficiency of these educational programs. This finding is similar to the results of numerous studies in this area (20).
Considering to the rural situation of the research environment, the researchers emphasized on social variables, especially, family and religious support. Community support has a great and effective role in the success of educational programs. These results have been confirmed in various studies aimed to promote self-care behaviors in different situations (16, 17).
The results showed that after the educational intervention, the mean scores of knowledge and attitude in the case group, unlike the control group, significantly increased, and a mean difference was seen between two groups. Based on the TPB model attitude has a direct impact on intention and intention has a direct impact on behavior. Therefore, if a woman has a positive attitude regarding EBF, she will have a stronger intention to accept the EBF as a healthy behavior. These findings were consistent with the results of similar studies (15). Similarly, our findings showed that after the intervention, the mean score of the norm in the experimental group, unlike the control group, increased significantly and the mean difference between the two groups was observed. This means that women perceived more social pressure to perform EBF as a health behavior (25, 26).
Also, after the intervention, the mean score of the perceived behavioral control in the experimental group improved significantly, and a mean difference was seen between the two groups. This means that women feel more perceived behavioral control associated with increasing the women’s perceptions of their ability to perform EBF and nourish their babies (12-19).
The construct of self-identity states that the perception or recognition of one's characteristics is a particular individual characteristic, especially in relation to social context. After the intervention, the mean score of self-identity increased in the intervention group, which means that women accepted the fact that breastfeeding was an important part of their motherhood role (15).
After the intervention, the women in the experimental group had a higher mean score of moral norms about breastfeeding than the control group. They believed that breastfeeding was a moral norm in the community, and mothers should breastfeed their babies (17).
The results show that perceived behavioral control has a significant effect on exclusive breastfeeding; however, attitude in some researches had a higher impact on behavior (26).
In this study, findings showed that intention was the only predictive variable for exclusive breastfeeding (Table 4). According to the assumptions of the theory of planned behavior, behavioral intention is the strongest predictor for behavior, which has been proven in numerous studies (19, 22-24). In contrast to these results in different studies, other model structures, including behavioral control, were the strongest predictor for behavior (25).
5.1. Limitations
Since it was not possible for the husbands to participate in training sessions, a training package was designed for them to compensate for this limitation.
5.2. Conclusions
Considering the findings of this study, after the intervention, the EBF in the intervention group was significantly increased due to the educational program. Therefore, it can be concluded that the extended theory of planned behavior is a suitable basic framework to promote the rate of EBF.
Also, the constructs of behavioral intention and perceived behavioral control were the significant predictors of EBF (Table 4). In this study, two new constructs could not predict the behavior, though they had a great impact on behavioral intention as a mediator of the behavior. As a result, considering the cultural differences, it is suggested that more studies be conducted.
