Investigators reported a notable shift in delivery patterns associated with documented receipt of personalized prenatal care in this single-center retrospective cohort. The cohort included 100 women with high-risk pregnancies who delivered between January 2020 and December 2023; after propensity-score matching, the analytic sample compared 50 women who received personalized care with 50 who received routine care. Vaginal delivery occurred in 62.0% of the personalized-care group compared with 40.0% of the routine-care group, an absolute difference of +22.0 percentage points (95% CI, +2.4 to +39.3). The authors reported an odds ratio of 2.45 (95% CI, 1.10–5.47; P = .028) for vaginal delivery favoring personalized care. Cesarean delivery rates moved in the opposite direction: 28.0% in the personalized group versus 48.0% in the routine-care group, with an absolute difference of −20.0 percentage points (95% CI, −37.1 to −1.0) and an OR of 0.42 (95% CI, 0.18–0.97).
These delivery-mode differences were observed in unadjusted, independent-group comparisons after matching. The study team emphasized that the design yields associations rather than causal inference. Clinical decision-making around mode of delivery can be multifactorial; the investigators did not report detailed intrapartum management protocols or the specific clinical reasons for cesarean delivery in each group in the abstract.
The study team observed lower preterm birth rates in the personalized-care arm. Preterm birth occurred in 10.0% of women who received personalized care versus 26.0% in the routine-care group, an absolute difference of −16.0 percentage points (95% CI, −30.7 to −0.8) with an OR of 0.32 (95% CI, 0.10–0.97; P = .037). One-minute and five-minute Apgar scores were reported as higher in the personalized-care group; the authors provided P values of .026 and .006, respectively.
Investigators also evaluated a composite adverse perinatal outcome defined as occurrence of at least one of preterm birth, low birth weight, neonatal asphyxia, or neonatal intensive care unit (NICU) admission. The composite occurred in 16.0% of the personalized-care group versus 36.0% of the routine-care group, an absolute difference of −20.0 percentage points (95% CI, −35.8 to −2.8). The reported OR for the composite outcome was 0.34 (95% CI, 0.13–0.88; P = .023).
The abstract did not report granular neonatal morbidities beyond the composite elements, nor did it provide individual counts for low birth weight, neonatal asphyxia, or NICU admissions. The study team limited conclusions to associations and noted limited power for several individual outcomes.
The authors described a propensity-score matching strategy intended to balance baseline risks between groups. Propensity scores incorporated maternal age, gravidity, parity, pre-pregnancy or early-pregnancy body mass index, gestational age at registration, gestational diabetes mellitus, hypertensive disorders of pregnancy, thyroid dysfunction, uterine scar, and history of adverse pregnancy outcomes. One-to-one nearest-neighbor matching without replacement was performed using a caliper of 0.2 standard deviations of the logit of the propensity score. All 100 women were retained after matching, yielding 50 matched pairs for analysis.
The abstract indicates investigators then used unadjusted independent-group comparisons to analyze outcomes. The authors did not report additional multivariable adjustment after matching in the abstract. Demographic and clinical distributions after matching were not detailed in the abstract text beyond the variables used to generate propensity scores.
Authors reported differences in selected nursing-compliance measures and end-of-care psychological scores favoring personalized care, according to the abstract. The specific nursing-compliance metrics, the psychological scales used, and the magnitude of those differences were not described in the abstract. The study team did not provide those instrument names or numeric results in the abstract text; therefore this summary does not extrapolate beyond what was reported.
The authors explicitly framed their findings as associations rather than causal effects. They noted several limitations in the abstract: a small sample size, potential residual confounding despite propensity-score matching, and limited statistical power for several individual outcomes. The study was single-center and retrospective; the abstract did not provide additional details on selection criteria beyond inclusion of women with high-risk pregnancies who delivered during the specified timeframe. The investigators also indicated no funding or conflicts of interest.
Unanswered questions that remain, based on the abstract, include the content and delivery mechanisms of the personalized prenatal program, the timing and intensity of interventions, and the causal pathways linking personalized prenatal care to mode of delivery and neonatal outcomes. The abstract does not report longer-term neonatal follow-up or detailed intrapartum management data. Future reports or the full-text article would need to clarify these elements to support interpretation of the observed associations.