The result showed the grandmothers’ knowledge mean scores were 9.32, lower than the mothers’ average scores of 10.83 [18]. Most grandmothers knew the benefits of breastfeeding, but about half did not know breastfeeding duration well, similar to other studies [24]. In some grandmothers’ options, water or other food should be complemented to meet infants’ nutrition or other requirements such as treating asthma [25]. At the same time, we found some scientific feeding knowledge had not been mastered by the partial grandmothers, such as the adequacy of colostrum, feeding on demand, not squeezing remaining breast milk, and judging the adequacy of feeding by the amount of urine. Our results showed that grandmothers with a high school education and above had the most breastfeeding knowledge, and the possible reason is that grandmothers with higher education backgrounds have better abilities or ways to accept breastfeeding knowledge. However, grandmothers with primary and below education have higher knowledge scores than junior or high school levels. This may be related to their better practical experience in breastfeeding or the small sample size, which needs further verification. Grandmothers’ higher income is related to more accurate breastfeeding knowledge, which may reflect the relationship between better socioeconomic levels and learning ability. The result indicated that breastfeeding lectures could improve grandmothers’ knowledge. However, in most cases, despite grandmothers' critical role within households, hospital staff often belittle their expertise and do not consider the design of health education [10]. Therefore, future clinical practice should consider these gaps in grandmothers’ breastfeeding health education.
This study found that grandmothers had a positive attitude towards breastfeeding, and grandmothers with higher educational levels and higher monthly incomes, who were not working, participated in breastfeeding literature and had a more excellent attitude towards breastfeeding. This might be because these grandmothers had a higher demand for the nutrition of infants and young children and had more time and scientific channels to understand breastfeeding advantages. Consistent with previous research, grandmothers’ education levels were positively related to breastfeeding attitudes [26]. The higher participation of unemployed female elders in supporting family breastfeeding might be related to the fact that older women are highly respected and occupy a higher status in the family [27, 28]. Besides, grandmothers had negative attitudes toward public breastfeeding. Over one-third of grandmothers thought mothers should not breastfeed in public, and over half of grandmothers agreed that formula feeding is the better choice if the mother returns to work. Thus, we may need to value breastfeeding importance, solve loopholes in current health education, and design multi-form plans to change grandmothers’ public breastfeeding attitudes in the future.
The mean score for mothers’ perception of grandmothers’ breastfeeding support scale was 27.77, with a positive level. However, some negative support was seen in the formula, which was the same as in previous research [21]. This phenomenon may be related to the prevalence of formula advertising and the lack of evaluation knowledge related to breastfeeding [29]. This study found that mothers with postgraduate and above, part-time working, and breastfeeding plans had more support from grandmothers, which may give them more time and skills to communicate and cooperate with their grandmothers [30]. In addition, the results showed that the higher the grandmother's scores on breastfeeding attitudes and knowledge, the higher the mothers' perceptions of family support for breastfeeding, which may be related to the grandmother's better knowledge and attitude to facilitate effective communication.
The results showed that mothers’ breastfeeding plans and support perception, grandmothers’ attitudes, and education level significantly impacted breastfeeding practices within six weeks postpartum. The theory of planned behavior states that attitudes and self-perceived behaviors influence a person's intention and actual behavior to perform a specific behavior [31]. Previous research also demonstrated that the mother’s feeding intention was independently related to the cessation of breastfeeding in the early postpartum period, which may be why breastfeeding plans affect practice [32, 33]. Attitudes toward the behavior affect the degree to which an individual evaluates and evaluates the behavior, while perceived behavioral control predicts the perceived difficulty in executing the behavior. Ong et al. [34] showed that female relatives (mother and mother-in-law) play an essential role in supporting new parents, and their breastfeeding attitude may affect the mother's self-perception, thus affecting breastfeeding practice. The study of Liu et al. [35] believed that highly educated grandmothers are related to the reduction of exclusive breastfeeding, which may reflect the relationship between better family socioeconomic status and preference for formula milk powder because formula milk powder is regarded as an indicator of higher socioeconomic status. Contrary to their research results, our study showed the education level of grandmothers had a positive effect on breastfeeding practice and further illustrated that the health management department had achieved staged victory in restricting the advertising of formula milk powder on infant feeding choices and vigorously supporting the practice of exclusive breastfeeding.
Some limitations must be considered when interpreting the result. First, grandmothers were recruited through convenience sampling in the postpartum ward, which might lead to research bias, mainly because grandmothers accompanying primiparous women might be more motivated to participate in breastfeeding. Second, 27.7% of grandmothers in this study lived separately from primiparas, which impacted mothers' perception of family support for breastfeeding and breastfeeding practices. However, the analysis showed that this effect was not statistically significant. Third, this study was conducted in an obstetrics and gynecology specialist hospital in Fujian Province, which may limit the generalizability of the results to other populations and settings.