Parental Socioeconomic Status as a determinant of breastfeeding : an analysis of the Korea National Health and Nutrition Examination Survey ( 2013 ~ 2017 )

Background: The importance of breastfeeding is well known. One of important factors affecting breastfeeding is socioeconomic status. We investigated the relationship between socioeconomic status and breastfeeding to promote future breastfeeding projects. Methods: Data were collected from the 2013 - 2017 Korea National Health and Nutrition Examination Survey (KNHANES). We evaluated the demographic information and parents’ socioeconomic status of 1,220 children under 60months. Results: A total 1,220 children were included in this study. Some of the socioeconomic factors were associated with breastfeeding. Mothers’ education level ( ≥ 13years: odd ratio[OR], 2.79; 95% condence interval [CI], 1.21-6.42), middle high of mother’s income level (OR, 2.30; 95% CI, 1.18-2.84), no smoking status (OR, 3.07; 95% CI, 1.28-7.36), body mass index(BMI) (< 25(kg/m 2 ): OR, 1.82; 95% CI, 1.12-2.95) were associated with breastfeeding (p<0.05). Also, fathers’ age (30s: OR, 4.88; 95% CI, 1.82-13.04), education level( ≥ 13years: OR, 7.94; 95% CI, 3.12-20.18) were associated with breastfeeding (p<0.05). After controlling for the confounding factors, mothers’ BMI, fathers’ age and educational level were statistically signicant. Conclusion: This study demonstrated that socioeconomic factors were associated with breastfeeding in Korea.

However, despite the recommendation to breastfeed exclusively for about the rst 6 months, less than 50% of infants were exclusively breastfed in the rst 3 months, and 24.9% were exclusively breastfed in the rst 6 months. 9 Between 1985 and 1995, the global rates of exclusive breastfeeding increased by 2.4% per year on average (increasing from 14% to 38% over 10 years). 10 The Global Breastfeeding Collective, led by United Nations Children's Fund and the WHO, set a target to increase the global rate of exclusive breastfeeding in the rst 6months of life: at least 50% by 2025 and at least 70% by 2030. 10,11 In Korea, the rates of breastfeeding at 6months of age increased to 60.8% in 2010 and 62.3% in 2013; however, it decreased to 55.6% in 2016. The Korean Ministry of Health and Welfare's Fourth National Health Plan aims to increase to 66.8% by 2020. 12 Hence, exploring the socioeconomic factors which affect breastfeeding are crucial for increasing breastfeeding rate and maintaining breastfeeding. Although the socioeconomic factors associated with breastfeeding have been broadly reported, scant studies have examined these variables in Korea. Therefore, we explored the relationship between socioeconomic factors and breastfeeding in Korea using data from the Korean National Health and Nutrition Examination Survey (KNHANES).

Data source
This study was conducted using the raw data from the KNHANES of the sixth (2013 ~ 2015) and the seventh term (2016~2017). The KNHANES is a statutory survey on people's health behavior, prevalence of chronic diseases, and food and nutrition practices. KNHANES was conducted in a three-year period from the rst (1998) to the third period (2005), and has since been reorganized into a year-round survey system, and has been conducted annually from the fourth term (2007)(2008)(2009)) to the present day.
KNHANES consisted of a cross-sectional survey composed of a health interview survey and a health examination survey, and a nutrition survey. All survey protocols were approved by the Korea Centers for Disease Control & Prevention (KCDC) Institutional Review Board. Written informed consent was obtained from all participants before the survey commencement.
In this study, we included 1,726 children under 60 months in the sixth and seventh terms. Of these, we excluded 506 participants who were no information of breastfeeding, did not provide socio-economic data. We included the 1,220 participants who had complete health interview survey in the nal analysis.

Study design
This study employed a cross-sectional design. All data are available from the KNHANES database (http://knhanes.cdc.go.kr/knhanes). We categorized data into two groups (non-breastfed, breastfed). The breastfeeding group was derived from the following questionnaires: "Have you been breastfeeding for at least a month?" These surveys included questions regarding sex, age, residence area, employment status, education level, incomes, parental smoking status, weight and height. Education level was divided into 3 categories: ≤ 9years, 10~12years, ≥ 13years. Income was classi ed into quartiles to determine income level (1: low, 2: middle low, 3: middle high, 4: high). Body mass index (BMI) was calculated as the ratio of weight to height 2 (kg/m 2 ).

Data analysis
Statistical Package for the Social Sciences (SPSS) complex sample procedures were employed since KNHANES data were collected through a representative, strati ed, and clustered sampling method.
Participants' characteristics are presented as a estimate and a 95% con dence interval. Differences between the breastfeeding and non-breastfeeding groups are reported using chi-square test. We conducted a logistic regression analysis to identify the relationships between factors and breastfeeding. We adjusted for mathers' education, income, smoking status, BMI, fathers' age, education level. Statistical analysis was conducted using SPSS version 21.0 (SPSS Inc. Chicago, IL, US). For all analyses, p-values were two-tailed, and a p value < 0.05 was considered signi cant.

Results
Participants' demographics are shown on Table 1. A total of 1,220 children were included in the study. Mean birth weight was 3.21kg.  To determine the association between breastfeeding and socioeconomic factors in all participants, logistic regression analysis was performed. The results are presented in Table 3

Discussion
Using data from the KNHANES, this study demonstrated that socioeconomic factors affect breastfeeding. Children whose mothers have a higher education level, middle high income, no smoking and obese status were associated with breastfeeding, as were children whose fathers had high education level and 30s age.
Multiple factors in uence breastfeeding. These factors were classi ed into four groups: demographic, biological, social, and psychological. 13 Demographic factors that affect the breastfeeding rate include race, maternal age, marital status, socioeconomic status, and education level. 13 In this study, we examined socioeconomic factors that affect breastfeeding.
Some studies showed that parental age did not affect breastfeeding. 14,15 However, several previous reports showed that the older the mothers were more likely to breastfeed than were younger mothers. [16][17][18] Oakely et al 19 reported that the younger the mother was, the less likely she was to breastfeed or to stop breastfeeding within 6 weeks. We posit that younger mothers lack knowledge and awareness of breastfeeding. In this study, most parents were in their 30s and most breastfed their infants; mothers' age did not affect breastfeeding.
Banu et al 20 stated that the higher the parents' education level, the higher the exclusive breastfeeding rate and longer duration of the exclusive breastfeeding. In this study, parents' education level was also related to breastfeeding. The higher the parents' education level, the more it affected breastfeeding. It is believed that the higher education level coincides with increased opportunities to access information about the bene ts of breastfeeding. In high-income countries, mother's education level is positively associated with higher breastfeeding rate. 16,17,19,[21][22][23][24][25] Even for mothers with a job, breastfeeding rate is estimated to be high, because the higher the level of education, the higher the occupational status, the better the working environment and the better knowing the bene ts of breastfeeding. In middle-and low-income countries, the relationship between breastfeeding and mothers' education level varies. Some studies have reported a negative relationship between mothers' education level and the breastfeeding rate owing to quick return to work. 18,26 Another study showed a higher parental education level was associated with a higher breastfeeding rate and exclusive breastfeeding. 27 We found a positive association between father's education level and breastfeeding. Similar nding has been reported by Flacking et al 28 , who found that the lower the fathers' education level, the lower the breastfeeding rate. We posit that this is related to family income. Because higher education often results in a better income, mothers may not need to work and can focus on parenting. However, other study showed that mothers'-but not fathers'-education level was related breastfeeding, 17 or that parental education level was not related to breastfeeding. 29,30 Wallby et al 31 reported that higher breastfeeding rate in low-income households. This is believed to be due to the fact that there are no economic condition to choose a different formula besides breast milk and no easy nding a job for mother. Victoria et al 4 stated that high-income, better-educated women breastfeed more commonly in high-income countries. Due to economic bene ts, the lower the household's gross income, the higher the breastfeeding in low-income and middle-income countries.
However, in other studies, household income did not affect breastfeeding. 15,29 In this study, there was no relationship between house income, the parental income and breastfeeding, however, the mothers of middle high income was related to breastfeeding.
Maternal job status is variable that has been associated with breastfeeding. Previous studies 14,18,24 revealed that full-time houseswives have higher breastfeeding rates than mothers with work and the shorter the time to return to work after childbirth, the shorter the breastfeeding period. It is estimated that they stop breastfeeding because they have less time to care for the child while working. However, we did not nd a signi cant association between breastfeeding and parental job status.
A consistent negative association between breastfeeding and maternal smoking has been well known in present study. 13,32 Weiser et al 33 showed that smoking during the postpartum period were associated with failure to initiate breastfeeding, also associated with weaning sooner. Nicotine increases dopamine secretion in the hypothalamus, thereby reducing prolactin levels. 34 For this reason, helping mothers quit smoking is bene cial to the health of infants and children as it helps prolong breastfeeding duration. In our study, signi cant negative association with breastfeeding was observed for maternal smoking.
Baker et al 35 suggested that maternal obesity can be considered as a risk factor to adverse breastfeeding outcome. There were some hypotheses trying to explain possible reasons why obese women are less likely to breastfeed; (1) large breast in obese women have been associated with breastfeeding practical di culties; 36 (2) excessive maternal adiposity may interfere with the development of the mammary glands; 37 (3) delayed lactogenesis and lower prolactin response to suckling in obese women. 38 Our result showed that obesity is associated with breastfeeding. But we were not clear on the possible reasons.
Importantly, our results shed light on some socioeconomic factors associated with breastfeeding; however, this study has some limitations. The utilized data were collected through self -report, and this is a source of recall bias which may have either underestimated or overestimated the relationship between the breastfeeding and socio-economic status. This study was conducted by breastfeeding status, it was not possible to investigate the effects of breastfeeding duration and type. Further, causal relationships cannot be con rmed owing to the cross-sectional design of the KNHANES. In addition to the factors used in this study, it is considered that studies on various social factors such as breastfeeding education, use rooming-in, use of postpartum care center, parental leave, and establishment of a breastfeeding room in the workplace are necessary.

Conclusion
This population-based study demonstrated that socioeconomic factors (parental education level, mothers' income, smoking and obese status and fathers' age) affect breastfeeding. To increase the breastfeeding rate and exclusive breastfeeding, further research should explore the reasons why parents stop breastfeeding. All of the participants in the KNHANES were informed that they had been randomly chosen to participate in the survey with the right to participate in the further analyses, and signed an informed consent form. As this was a cross-sectional study that used and analyzed data from KNHANES(http://knhanes.cdc.go.kr/knhanes/), ethical approval was not required. Flow chart of the sample selection process of the study population