BackgroundThe benefits of breastfeeding, to infant and mother, are supported by a comprehensive body of evidence. However, in China, only a small proportion of women achieve the breastfeeding recommendation from WHO. Previous studies in high-income countries have found a positive association between socioeconomic status and initiation and duration of breastfeeding. This association might differ between societies and be affected by rapid social and economic development, such as has occurred in China over the past decade. The purpose of this study was to investigate the role of socioeconomic status on breastfeeding initiation and duration in China. MethodsIn this analysis we used data from the China Family Panel Studies (CFPS), funded by the 985 Program of Peking University and undertaken by the Institute of Social Science Survey of Peking University. CFPS is a nationally representative and longitudinal household survey that has been collecting individual, family, and community information annually in China since 2010. CFPS has surveyed 15 000 selected households in 25 provinces or directly governed municipalities. Our analysis included all surveyed children who were born between 2010 and 2014, and excluded children with missing breastfeeding information. We obtained data for initiation of breastfeeding (ie, breastfed or non-breastfed), and duration of breastfeeding time, expressed in months. Socioeconomic status was measured using household income per capita, parental years of education, and parental occupational status. We used Kaplan-Meier time-to-event analysis to compare duration of breastfeeding according to socioeconomic status variables, a logistic regression model to analyse the association between socioeconomic status and initiation of breastfeeding, and a Cox proportional hazards model to analyse breastfeeding duration and socioeconomic status. Sex of infant, birthweight of infant, birthplace, nationality, type of residence, age of parents, residential region, marriage status of mother, parity of mother, and birth year were adjusted for in the logistic and Cox model. CFPS was undertaken according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human participants were approved by the ethics committee of Peking University. Written informed consent was obtained from all participants. FindingsOur sample consisted of 2938 children. 2658 (90%) infants had ever breast-fed, and the mean duration of breastfeeding was 8·66 months (SD 6·15). After adjusting for potential confounders, we found that socioeconomic status variables were not significant predictors for breastfeeding initiation, but that low birthweight babies were less likely to be breastfed than normal weight babies (odds ratio 0·272, 95% CI 0·18–0·42; p<0·0001). As for breastfeeding duration, mothers with a high school or higher education were more likely to breastfeed for longer than those with a middle school or lower education (hazard ratio [HR] 1·15, 95% CI 1·02–1·29; p=0·021), as were infants with fathers with the highest occupation level versus those with fathers with the lowest occupation level (HR 1·29, 95% CI 1·04–1·59, p=0·019). Mother's occupation had an inverted U-shaped relationship with breastfeeding duration (HRs for low, high, and highest occupation level vs lowest occupation level were 1·17 [95% CI 1·02–1·33], p=0·023; 1·29 [1·08–1·54], p=0·005; and 1·11 [0·87–1·42], p=0·408, respectively). InterpretationThe finding that socioeconomic status in China was not a predictor of breastfeeding initiation is different from that reported in high-income countries. Comprehensive efforts to promote breastfeeding practices should be targeted towards women who are less well educated, and those in both the lowest and highest occupation levels, because we found that these variables were associated with shorter breastfeeding duration. FundingSichuan University (skqx201401 and 2015SCU04A19).