A cross-sectional study of 460 postpartum women in Chengdu, China, found that 38.5% experienced breastfeeding difficulties between 24 and 96 hours after delivery. Perceived insufficient milk was the most common primary problem, accounting for 63.8% of affected participants, followed by nipple-related difficulties at 11.3%. Multivariable analysis identified several factors independently associated with higher odds of early difficulties: breastfeeding fewer than six times per day, gestational diabetes mellitus, a pre-pregnancy body mass index of 24.0–27.9 kg/m², and elevated postpartum depressive symptoms. Stronger family support and higher LATCH scores for feeding effectiveness showed inverse associations. Because the research was observational, these results identify statistical relationships rather than proving that any factor caused the feeding barriers.
The study, conducted at a tertiary specialist hospital between July 2023 and November 2024, used convenience sampling to recruit women who intended to breastfeed or were already doing so. Participants were assessed from 24 to 96 hours postpartum, a window that includes both the period before full secretory activation and the transitional phase of copious milk production. Five women with incomplete questionnaires were excluded from the final dataset, which comprised 177 participants with breastfeeding difficulties and 283 without them. Two nurses certified by the International Board Certified Lactation Consultants collected questionnaire data, reviewed medical records and observed feeding sessions. The assessment window was deliberately broader than secretory activation alone, so the outcome was described as early postpartum breastfeeding difficulty rather than delayed lactogenesis.
Difficulty was classified as present if a mother had at least one documented clinical barrier: perceived insufficient milk accompanied by substantial anxiety or non-medically indicated formula supplementation; nipple or breast problems such as severe pain, trauma or engorgement; or neonatal and interaction-related factors such as inability to maintain a latch, persistent crying at the breast or poor weight gain requiring intervention. Participants without these barriers were classified as having no difficulty. The researchers also measured family support with a nine-item checklist covering practical assistance and verbal encouragement, breastfeeding confidence with the 14-item Breastfeeding Self-Efficacy Scale-Short Form, depressive symptoms with the Edinburgh Postnatal Depression Scale and feeding technique with the LATCH tool, which rates latch, swallowing, nipple type, maternal comfort and positioning on a 0–10 scale.
Among women classified as having difficulties, 113 had perceived insufficient milk as their primary manifestation, 20 had nipple-related problems, 12 had neonatal health issues, 11 had inadequate infant satiety or poor weight gain, seven had other breast-related problems, five lacked family support, five were preparing to return to work and four perceived the nutritional content of their milk as inadequate. The distribution was not restricted to one clinical mechanism. A mother could therefore experience a combination of physical, psychological and infant-related barriers, while each participant was assigned one mutually exclusive category for the reported classification. The LATCH tool also captured effective feeding, with higher scores interpreted as better positioning, latch and milk transfer.
The clearest adjusted association involved feeding frequency. Compared with women breastfeeding eight to 12 times per day, those doing so fewer than six times had 3.434 times the odds of experiencing difficulties, with a 95% confidence interval of 2.060–5.725. Gestational diabetes was associated with 1.819 times the odds, or a 95% confidence interval of 1.042–3.177, while the BMI category of 24.0–27.9 kg/m² was associated with 2.484 times the odds, with a 95% confidence interval of 1.093–5.649. The highest BMI category, at least 28.0 kg/m², had a larger point estimate of 3.471 but a wide confidence interval of 0.776–15.540 and was not statistically significant. Elevated depressive symptoms, defined by an EPDS score of at least 10, were associated with 3.365 times the odds of difficulties, with a 95% confidence interval of 1.545–7.329. Overall, 52 women screened positive for elevated depressive symptoms.
Two protective associations also remained after adjustment for maternal age, delivery mode and parity, among other variables. Each one-point increase in the family-support score was associated with lower odds of difficulty, with an odds ratio of 0.911 and a 95% confidence interval of 0.861–0.965. Expressed as a five-point difference, this corresponded to 37.3% lower odds, with an odds ratio of 0.627 and a 95% confidence interval of 0.473–0.837. Each one-point increase in the LATCH score was associated with an odds ratio of 0.646, with a 95% confidence interval of 0.572–0.729. A clinically meaningful two-point increase corresponded to 58.3% lower odds, with an odds ratio of 0.417 and a 95% confidence interval of 0.327–0.531. Breastfeeding self-efficacy did not independently reach statistical significance in the primary model, although it showed a favorable trend.
The statistical model had an area under the receiver operating characteristic curve of 0.829 and a Nagelkerke pseudo-R² of 0.411, with a nonsignificant Hosmer–Lemeshow goodness-of-fit result of P = 0.227. Because the LATCH score overlaps conceptually with the clinical definition of breastfeeding difficulty, the researchers repeated the model without it. The remaining risk factors—less frequent breastfeeding, elevated BMI, gestational diabetes and elevated depressive symptoms—remained associated with difficulty in this sensitivity analysis. The revised model had an area under the curve of 0.764 and Nagelkerke pseudo-R² of 0.280. This consistency supported the persistence of the main associations, although removing LATCH reduced the model’s apparent discriminatory performance.
The findings point toward a multidimensional early assessment combining maternal health, feeding technique, psychological state and family support, but they do not establish that a bundled intervention would improve breastfeeding outcomes. Frequent effective suckling and effective milk removal are biologically linked to the local supply-and-demand regulation of milk production, while a deep, comfortable latch can improve transfer and reduce mechanical nipple trauma. The observational findings are compatible with those mechanisms, but the study did not measure prolactin, progesterone, oxytocin or milk-removal efficiency. The direction of association between depressive symptoms and breastfeeding difficulty also remains uncertain: mood could impair feeding confidence and routine, whereas unsuccessful early feeding could itself increase distress. Future work therefore requires larger multicentre prospective cohorts, objective endocrine measures, detailed infant oral-motor assessment and randomized trials of targeted support.
Single-centre convenience sampling is the study’s clearest boundary on interpretation. All participants came from one tertiary specialist hospital in Chengdu, where maternity-care practices, rooming-in arrangements and access to lactation support may differ from those in other regions or hospitals. Self-reported maternal information and observational scales added further uncertainty, while five incomplete cases were removed rather than imputed. The sample also excluded women with severe obstetric complications, medical contraindications to breastfeeding, certain congenital anomalies, serious psychiatric or cognitive disorders and medications known to inhibit lactation, limiting its reach beyond relatively stable postpartum populations. Even with those limitations, the results identify where early clinical attention may be most productive: frequent feeding and skilled latch, alongside attention to metabolic health, depressive symptoms and practical support from family members, warranting evaluation in prospective intervention studies.
Subject of Research: Clinical factors associated with breastfeeding difficulties during the first 96 hours postpartum in China
Article Title: Factors associated with breastfeeding difficulties in the early postpartum period in China: a cross-sectional study
Article References: Lei, F., Zhou, S., Liu, Y., Xiang, J., Li, N., & Liu, Y. (2026). Factors associated with breastfeeding difficulties in the early postpartum period in China: a cross-sectional study. Frontiers in Public Health, 14, Article 1923736. https://doi.org/10.3389/fpubh.2026.1923736
Image Credits: AI Generated
DOI: 10.3389/fpubh.2026.1923736
Keywords: breastfeeding difficulties, early postpartum period, China, latch assessment, family support, postpartum depression, gestational diabetes, maternal BMI, lactation support, Factors, associated, breastfeeding
News Source: Ophelia Keating. (October 20, 2026). Study links breastfeeding barriers to health, skills and support. Scienmag.
