A randomized controlled trial of a health literacy–based digital intervention to promote postpartum recovery and self-care

Abstract

To evaluate a six-week Postpartum Comprehensive Health Intervention Program (PCHIP), a health literacy-based digital intervention designed to promote early postpartum self-care and selected recovery outcomes. This pre-registered, two-arm, single-blind randomized controlled trial was conducted at Changzhou Second People’s Hospital, Changzhou, China. Participants were primiparous women (mean age 29.75 years) recruited from January 25 to February 15, 2025 and enrolled within the first three days postpartum. Eighty-eight participants were randomized 1:1 using a computer-generated sequence, with allocation concealed in sealed opaque envelopes, to PCHIP plus standard education or standard education alone. PCHIP began on postpartum day 3 and lasted six weeks. Primary outcomes were eating behaviors, physical activity, and breastfeeding practices; health literacy and recovery indicators were secondary outcomes. Complete-case analyses included 82 women. ANCOVA was used for continuous outcomes with baseline values, weight-related outcomes were adjusted for baseline postpartum weight, and Fisher’s exact test was used for exclusive or predominant breastfeeding. Eighty-two participants completed follow-up (PCHIP: 40/44; control: 42/44). Baseline-adjusted analyses showed lower pain (AMD = -0.698, 95% CI: -1.339 to -0.057), depression (AMD = -1.313, 95% CI: -2.398 to -0.229), anxiety (AMD = -1.357, 95% CI: -2.543 to -0.170), emotional eating (AMD = -4.421, 95% CI: -8.177 to -0.666), and external eating (AMD = -4.043, 95% CI: -7.957 to -0.129), and higher health literacy (AMD = 2.714, 95% CI: 0.032 to 5.397) and physical activity (AMD = 0.969, 95% CI: 0.227 to 1.711) in the PCHIP group. Restrained eating showed a favorable but non-significant trend. Weight-related analyses adjusted for baseline postpartum weight showed greater weight reduction in the PCHIP group (AMD = 2.527 kg, 95% CI: 0.835 to 4.218). Outcomes measured only at 6 weeks suggested shorter red lochia duration (MD = -1.224 weeks, 95% CI: -1.811 to -0.638), better selected pelvic floor muscle indicators, and more frequent exclusive or predominant breastfeeding (risk difference = 22.6%, 95% CI: 2.4% to 42.8%) in the PCHIP group. No adverse events were reported. PCHIP may support early postpartum self-care and selected recovery outcomes, but findings should be interpreted cautiously because of the single-center design, small sample, complete-case analysis, short follow-up, and multiple outcomes. Larger multicenter trials with longer follow-up are needed. The trial was prospectively registered with the Chinese Clinical Trial Registry (ChiCTR2500096473) on January 24, 2025.

Introduction

The postpartum period is a critical window for women’s recovery, yet many face persistent challenges that undermine maternal well-being and infant outcomes. Physiological problems are common: 20%–25% of women fail to return to pre-pregnancy weight within six months, and 10%–20% retain more than 5 kg, elevating long-term risks of obesity, diabetes, and metabolic syndrome^1,2. Unhealthy dietary behaviors and insufficient physical activity remain prevalent, often reinforced by sociocultural norms such as zuo yuezi (“sitting the month”), which prioritizes rest but may encourage sedentary lifestyles and imbalanced diets^3,4.

In parallel, psychological symptoms are widespread. Postpartum depression affects 13% globally and up to 30% in China, frequently accompanied by anxiety⁵. Fatigue and pain further limit daily functioning, breastfeeding continuation, and caregiving, collectively impairing quality of life and mother–infant bonding⁶.

Despite recognition of these problems, current interventions remain insufficient. Conventional postpartum education is knowledge-based but lacks personalization, feedback, and skill-building^7,8. Digital programs show potential yet struggle with sustained engagement, with over 50% attrition within two weeks⁹. Crucially, most approaches overlook health literacy—a determinant of self-care. In China, fewer than 40% of postpartum women demonstrate adequate health literacy, particularly in diet, physical activity, and information evaluation¹⁰. Low health literacy increases vulnerability to misinformation (e.g., “exercise is harmful in early postpartum”) and hinders evidence-based behavior adoption¹¹.

Health literacy not only influences dietary and activity behaviors but also shapes psychological resilience, self-efficacy, and the capacity to manage fatigue and pain¹². Enhancing functional, interactive, and critical health literacy may thus provide a unifying pathway to improve multiple outcomes¹³. However, mechanisms linking health literacy to postpartum recovery remain underexplored.

To address this gap, we developed the Comprehensive Health Behavior Intervention Program-Postpartum (PCHIP). The new concept of this manuscript is a health literacy-driven digital postpartum care model that treats health literacy not only as knowledge acquisition but also as an operational mechanism for self-care. Specifically, PCHIP integrates Orem’s Self-Care Theory, Nutbeam’s functional-interactive-critical health literacy framework, behavior change techniques, family-supported practice, and WeChat-based feedback to convert postpartum information into feasible daily behaviors. This approach extends conventional postpartum education by linking literacy enhancement, behavior execution, and early recovery support within one structured digital intervention.

Background

Theoretical framework

This study is conceptually grounded in Orem’s Self-Care Theory and Nutbeam’s Health Literacy Model, which together provide a complementary framework for understanding how postpartum interventions foster sustained health behavior change^13,14.

Orem’s Self-Care Theory defines self-care behaviors as outcomes of self-care agency—the individual’s capacity to mobilize resources to meet therapeutic self-care demands. In the postpartum context, such behaviors include dietary regulation, physical activity, and breastfeeding. When self-care agency is insufficient, self-care deficits emerge, requiring supportive–educative nursing interventions¹⁴. The present intervention builds on this foundation by aiming to strengthen women’s self-care agency through structured enhancement of health literacy.

Nutbeam’s Health Literacy Model extends this framework by conceptualizing literacy as a multidimensional construct comprising functional, interactive, and critical domains¹³. Functional literacy equips women to comprehend essential health information (e.g., dietary guidelines, safe exercise practices). Interactive literacy enables effective communication with healthcare providers and family members, thereby improving adherence to recommended practices. Critical literacy empowers women to evaluate and adapt health information, facilitating the reconciliation of traditional practices such as zuo yuezi with evidence-based recommendations.

Integrating these models, health literacy can be viewed as the operational pathway through which self-care agency is actualized. Enhancing literacy across functional, interactive, and critical domains enables postpartum women to make informed decisions, sustain dietary and physical activity behaviors, and manage breastfeeding more effectively (Fig. 1).

Fig. 1.

Theoretical substruction: Integration of self-care theory and health literacy.

Hypothetical model

Building on this theoretical integration, the study posits a causal framework (Fig. 2) in which the Comprehensive Health Behavior Intervention Program–Postpartum (PCHIP) functions as the independent variable, health literacy serves as the mediating construct, and postpartum self-care behaviors represent the dependent outcomes. Through tailored education and interactive support, the intervention is expected to enhance health literacy, which in turn drives the adoption and maintenance of health-promoting behaviors. This model positions health literacy simultaneously as an intervention target and a mechanism of behavior change.

Fig. 2.

Hypothetical model for the Comprehensive Health Behavior Intervention Program-Postpartum (PCHIP).

Methodology

Study design

This study was a two-arm, parallel-group randomized controlled trial (RCT) conducted among postpartum women in urban China. Participants were randomized in a 1:1 ratio to either the intervention group, receiving the Comprehensive Postpartum Health Management Program, or the control group, receiving routine postpartum care. The trial was guided by Orem’s Self-Care Theory and Nutbeam’s Health Literacy Model, with intervention content aligned with national postpartum guidelines^13–15.

Participants and setting

The study population consisted of primiparous postpartum women who gave birth at Changzhou Second People’s Hospital and attended postpartum care at the Maternity Clinic. Participants were recruited between January 25, 2025 and February 15, 2025. Eligible women were screened and enrolled within the first three days postpartum through systematic random sampling using hospital appointment numbers. The intervention began on postpartum day 3 and lasted for six weeks.

Inclusion criteria were as follows: age 18–35 years, primiparous, pre-pregnancy BMI 18–28 kg/m², vaginal delivery, enrollment within the first three days postpartum, smartphone access, and at least one family member involved in meal preparation. Exclusion criteria were uncontrolled hypertension, thyroid or cardiovascular diseases, cancer, severe gastrointestinal disorders, postpartum complications (e.g., hemorrhage, anemia, or lacerations), severe mental illness, substance abuse, smoking, or inability to adhere to the intervention protocol.

Eligibility screening was performed by trained midwives and obstetricians.

Sample size

The sample size was estimated in relation to the prespecified behavioral primary outcomes of the present trial, including eating behaviors, physical activity, and breastfeeding practices. Because few prior randomized trials have evaluated health literacy-based digital interventions targeting multiple postpartum self-care behaviors in early postpartum women, a previous randomized controlled trial of diet and exercise intervention was used as a pragmatic reference for estimating a medium intervention effect on behavior-related maternal outcomes¹⁶. Assuming an effect size of d = 0.63, alpha = 0.05, and power = 0.80, a minimum of 80 participants (40 per group) was required. Allowing for approximately 10% attrition, the final sample size was set at 88 participants (44 per group). The study was not powered to draw definitive conclusions for each individual secondary recovery outcome.

Randomization and blinding

Randomization was performed using a computer-generated sequence (SAS software), with allocation concealed in sealed, opaque envelopes. Due to the behavioral nature of the intervention, participant blinding was not feasible. However, outcome assessors (midwives) were blinded to group allocation. Separate healthcare teams provided intervention and routine care to minimize contamination.

Intervention: postpartum comprehensive health intervention program

The postpartum comprehensive health intervention program was developed to promote maternal recovery by targeting dietary behavior, physical activity, breastfeeding practices, and pelvic floor rehabilitation. To ensure the intervention was both theoretically grounded and behaviorally specific, the design was guided not only by Nutbeam’s Health Literacy Model but also systematically mapped to the Behavior Change Technique (BCT) Taxonomy (version 1) developed by Michie et al.¹⁷. The BCT framework provides a standardized, evidence-based system for identifying and applying discrete components that drive behavior change, such as goal setting, self-monitoring, feedback, and social support.

Integrating the BCT framework offered several advantages. First, it allowed for the operationalization of health literacy constructs—transforming abstract literacy skills (functional, interactive, and critical) into concrete, actionable strategies that participants could practice. Second, it enhanced intervention transparency and reproducibility, enabling precise reporting and replication in other contexts. Third, it facilitated systematic tailoring, as specific BCTs could be adapted to individual postpartum needs and cultural norms.

The intervention incorporated three functional components. First, tailored information was delivered through digital manuals and interactive videos, providing participants with practical knowledge about balanced dietary plans, portion control, nutrient-rich food choices, and safe postpartum exercises such as stretching, yoga, and walking. The content emphasized both the long-term risks of excessive caloric intake and physical inactivity and the psychological benefits of healthy practices. Second, tailored communication was implemented through twice-weekly online consultations on WeChat (two sessions per week, 12 sessions in total over six weeks; approximately 20–30 min per session), which facilitated feedback on dietary and activity behaviors, weight management, and breastfeeding. These consultations also encouraged the involvement of family members, thereby strengthening social and practical support for lifestyle change. Third, tailored education focused on skill-building and motivation. Participants accessed video demonstrations illustrating meal preparation, portion management, home-based exercises, and pelvic floor muscle training. Automated digital reminders reinforced adherence and emphasized the benefits of sustained behavior change.

Based on these functional components, several core elements were integrated into the program, including stepwise guidance for behavior execution, video-based demonstrations of dietary and physical activity practices, education on the short- and long-term consequences of health behaviors, mobilization of family support, structured goal setting and progress monitoring, automated prompts and cues to encourage compliance, and positive reinforcement through personalized feedback. These components were systematically aligned with corresponding BCTs, as detailed in Table 1, to ensure consistency between theoretical principles and intervention delivery.

Table 1.

Intervention components and corresponding behavior change techniques.

Component type	Specific content	BCT code	Behavior change technique
Tailored Information/Guidance	Videos and manuals providing balanced diet planning, portion control, safe postpartum exercises	4.1	Instruction on how to perform a behavior
	Explanation of health risks of unhealthy behaviors (weight retention, chronic disease)	5.1	Information about health/environmental consequences
	Psychological benefits of healthy behaviors (stress reduction, mood improvement)	5.2	Information about emotional consequences
Tailored Communication/Social Support	Family involvement in meal preparation and exercise routines; encouragement of family support	3.1, 3.2	Social support (unspecified and practical)
	Twice-weekly WeChat consultations (12 sessions total; approximately 20–30 min per session) providing personalized feedback and monitoring	6.1	Feedback and monitoring
Tailored Education/Demonstration	Video tutorials on meal preparation and home-based exercises	4.3	Demonstration of behavior
	Automated digital reminders to reinforce adherence	7.1	Prompts/cues
Core Reinforcement Elements	Emphasis on short- and long-term benefits of behavior change	9.3	Information about behavior change benefits
	Goal setting and digital tracking of behavioral targets	1.1	Goal setting (behavior)
	Recognition of progress through verbal praise and incentives	10.4	Social reward

The program was implemented over six weeks, beginning on postpartum day 3. During weeks 1–2, participants received digital manuals and ten foundational educational videos covering dietary guidance, physical activity, and breastfeeding practices. Participants were asked to watch each foundational video at least once, and repeated viewing was encouraged according to individual needs. From week 3, an additional set of pelvic floor rehabilitation videos was introduced to provide gradual, safety-oriented instructions for pelvic floor muscle training. Participants were asked to watch each assigned pelvic floor video at least once before practicing and could review the videos repeatedly during the intervention. No additional minimum viewing time or required number of repeated views was imposed beyond completion of the assigned videos. Throughout the intervention period, participants received twice-weekly online consultations with trained providers, who monitored weight, diet, activity levels, breastfeeding practices, and exercise performance and offered tailored feedback to address barriers and reinforce adherence. The intervention concluded with a 6-week post-test assessment.

This structured, six-week program operationalizes the new concept of literacy-to-behavior conversion by combining concise digital education, interactive feedback, family support, and tailored behavioral reinforcement in early postpartum care.

Outcomes and measurements

Demographic and socioeconomic characteristics

Collected via structured questionnaires at baseline (age, marital status, education, income, pre-pregnancy BMI, gestational weight gain, postpartum duration, ethnicity, religion).

Primary outcomes

Eating behaviors: The Dutch Eating Behavior Questionnaire (DEBQ) was used to assess restrained, emotional, and external eating (Cronbach’s alpha = 0.75–0.90)¹⁸.

Physical activity: The Chinese version of the Pregnancy Physical Activity Questionnaire (PPAQ-C) was used to assess physical activity, with scores expressed as metabolic equivalent hours per week (MET-h/week), ICC > 0.80^19,20.

Breastfeeding practices: Breastfeeding status was self-reported at follow-up and categorized according to feeding pattern, including formula only, mixed feeding, partial breastfeeding, predominant breastfeeding, and exclusive breastfeeding²¹.

Secondary and mechanistic outcomes

Health literacy: Health literacy was assessed as a secondary/mechanistic outcome using the Health Literacy Survey-Short Form 12 (HLS-SF12), with an index score ranging from 0 to 50²⁴.

Psychological outcomes: Depression was assessed using the Edinburgh Postnatal Depression Scale (EPDS), and anxiety was assessed using the Generalized Anxiety Disorder Scale (GAD-7)^22,23.

Fatigue and pain: Fatigue was assessed using the Postpartum Fatigue Scale, and pain was assessed using the Numeric Rating Scale (NRS, 0–10)^25,26.

Recovery-related outcomes: Postpartum weight change, duration of red lochia, and pelvic floor muscle function were assessed as secondary recovery outcomes. Duration of red lochia was recorded in weeks. Weight change at 6 weeks postpartum was defined as the difference between maternal body weight measured within the first three days postpartum and maternal body weight measured at the 6-week follow-up²⁷; a positive value indicated weight reduction after delivery. For weight-related outcomes, maternal body weight measured within the first three days postpartum was treated as the baseline postpartum weight and was considered as a relevant baseline covariate in the adjusted analysis.

Pelvic floor muscle function was assessed at the 6-week follow-up using standardized pelvic floor muscle testing indicators, including resting muscle score, fast-twitch muscle score, 10-second slow-twitch muscle score, 60-second slow-twitch muscle score, and post-resting score.

Because pelvic floor muscle function and several recovery indicators were measured only at the 6-week postpartum assessment, baseline-adjusted analysis was not applicable for these outcomes.

Adverse events

Adverse events related to physical activity, breastfeeding, or intervention participation were monitored and recorded at each follow-up.

Data collection schedule

Data collection was conducted at baseline (within the first three days postpartum) and at the 6-week postpartum follow-up. Figure 3 illustrates the corrected schedule of participant enrollment, randomization, intervention delivery, and assessment activities.

Fig. 3.

Sequence of participant enrollment, intervention, and assessment activities after correction of the recruitment window and follow-up schedule.

At baseline, trained researchers blinded to group allocation collected sociodemographic and obstetric characteristics, psychological assessments, health literacy, dietary behaviors, physical activity, breastfeeding practices, maternal weight, and infant growth metrics.

At the 6-week post-test, the same behavioral and psychological measurements were repeated when applicable. Recovery-related outcomes, including pelvic floor muscle function, duration of red lochia, weight change, and adverse events, were also recorded. In the intervention group, adherence and process measures were monitored through WeChat logs and participant self-monitoring records.

Treatment fidelity program

A comprehensive fidelity program was established to ensure the consistency, quality, and reliability of the Postpartum Comprehensive Health Management Intervention. The program followed the guidelines of the NIH Behavior Change Consortium (BCC), covering intervention design, provider training, delivery, and evaluation.

Intervention providers were certified midwives or maternal–child health nursing specialists who completed additional training in postpartum nutrition, communication skills, and the latest Chinese Postpartum Dietary Guidelines²⁸. Only those who passed standardized assessments were authorized to deliver the intervention. Weekly quality checks and feedback ensured adherence to protocol, and additional training was provided when performance did not meet standards.

The intervention was delivered using a structured checklist to guarantee complete and accurate coverage of all components. Fidelity monitoring included random observations and video recordings of sessions. Participant adherence was supported through self-monitoring logs, automated reminders, and follow-up calls for missed sessions. Incentives such as free postpartum fitness or nutrition evaluations were offered to encourage participation. Brief interviews were also conducted to assess participants’ understanding and satisfaction, and tailored support was provided when needed.

Human subject protection

This study was approved by the Clinical Research Ethics Committee of the Affiliated Changzhou No. 2 People’s Hospital of Nanjing Medical University on January 10, 2025 ([2024]KY311-01). The trial was prospectively registered with the Chinese Clinical Trial Registry on January 24, 2025 (ChiCTR2500096473), before enrolment of the first participant. Participant recruitment began on January 25, 2025 and was completed on February 15, 2025. All procedures complied with the principles of the Declaration of Helsinki.

Eligible participants were fully informed about the study’s purpose, procedures, potential risks and benefits, confidentiality measures, and their rights to withdraw at any time. Written informed consent was obtained from all participants prior to enrollment.

Data analysis

All analyses were conducted using IBM SPSS Statistics version 28. Descriptive statistics were used to summarize baseline characteristics. Continuous variables were presented as means with standard deviations (SD) if normally distributed, or as medians with interquartile ranges (IQR) if skewed. Categorical variables were expressed as frequencies and percentages.

The analysis population was revised and clearly defined as a complete-case dataset including participants with available 6-week follow-up data. For continuous outcomes assessed at both baseline and 6 weeks postpartum, between-group effects were estimated using analysis of covariance (ANCOVA), with the 6-week outcome value as the dependent variable, group allocation as the independent variable, and the corresponding baseline value as a covariate. Adjusted mean differences with 95% confidence intervals were reported.

For weight-related outcomes, baseline postpartum weight was included as a covariate because weight change was derived from maternal weight measured within the first three days postpartum and maternal weight measured at the 6-week follow-up. For recovery-related outcomes measured only at 6 weeks postpartum and without corresponding baseline measurements, such as pelvic floor muscle function and duration of red lochia, baseline-adjusted ANCOVA was not applicable. These outcomes were analyzed using unadjusted between-group comparisons, and mean differences with 95% confidence intervals were reported. Breastfeeding status was summarized descriptively and compared between groups using Fisher’s exact test. No formal adjustment for multiple comparisons was applied; therefore, findings for secondary outcomes should be interpreted as exploratory. Statistical significance was set at a two-tailed p-value < 0.05.

Bias control and data quality assurance

Several strategies were used to reduce bias and potential confounding. Random allocation and sealed opaque envelopes were used to reduce selection bias; outcome assessors were blinded to group allocation; eligibility criteria were prespecified to reduce clinical heterogeneity; and the two groups were comparable at baseline. Standardized, validated instruments were used for psychological, literacy, physical activity, eating behavior, and pain outcomes, and trained midwives or maternal-child health nursing specialists followed structured assessment and intervention checklists. Data were checked for range consistency and implausible values before analysis. The relatively large SDs observed for some 6-week outcomes were interpreted as reflecting expected inter-individual variability in early postpartum recovery rather than, by themselves, evidence of invalid data. To improve interpretability, we reported effect estimates with 95% confidence intervals and used baseline-adjusted ANCOVA when corresponding baseline values were available.

Results

A total of 88 women were randomized, with 44 allocated to the intervention group and 44 to the control group. At the 6-week postpartum follow-up, 82 participants completed post-intervention assessments, including 40 in the intervention group and 42 in the control group. Six participants did not provide 6-week follow-up data (four in the intervention group and two in the control group). The primary analysis was therefore conducted as a complete-case analysis among 82 participants. The corrected participant flow is presented in Fig. 4.

Fig. 4.

Flowchart of participant enrollment, allocation, follow-up, and complete-case analysis in the Postpartum Comprehensive Health Intervention Program (PCHIP) trial.

At baseline, there were no statistically significant differences between the intervention and control groups in terms of demographic, clinical, psychological, or behavioral variables (all p > 0.05), indicating good comparability between the two groups (Tables 2 and 3). All participants were of Han ethnicity and reported no religious affiliation, ensuring population homogeneity and minimizing potential cultural confounders.

Table 2.

Baseline demographic and clinical characteristics of participants (N = 88).

Variable	Control (n = 44)	Intervention (n = 44)	t/χ²	p value
Age, (years)	29.64 ± 3.44	29.86 ± 4.39	−0.272	0.786
Gestational weeks at delivery (weeks)	39.38 ± 1.24	38.98 ± 1.18	1.464	0.147
Pre-pregnancy BMI (kg/m²)	22.15 ± 3.13	22.94 ± 3.06	−1.193	0.236
Gestational weight gain (kg,)	14.56 ± 5.58	17.36 ± 9.57	−0.722	0.472
Postpartum weight (kg)	68.12 ± 10.78	67.86 ± 9.27	0.122	0.903
Neonatal birth weight (kg)	3.29 ± 0.35	3.17 ± 0.56	0.840	0.403
Education level, n (%)			2.618	0.454
Below college	5 (11.4)	6 (13.6)
College diploma	14 (31.8)	13 (29.5)
Bachelor’s degree	20 (45.5)	19 (43.2)
Master’s degree or above	5 (11.4)	6 (13.6)
Annual household income, n (%)			5.151	0.272
≤100,000	7 (15.9)	6 (13.6)
100,001–200,000	15 (34.1)	16 (36.4)
200,001–300,000	13 (29.5)	14 (31.8)
≥400,000	9 (20.5)	8 (18.2)

Data are presented as mean ± standard deviation (SD) for continuous variables and n (%) for categorical variables. p-values were calculated using independent t-tests for continuous variables and χ² or Fisher’s exact tests for categorical variables.

Table 3.

Baseline psychological, behavioral, and literacy characteristics of participants (N = 88).

Variable	Intervention (n = 44)	Control (n = 44)	t	p-value
Pain	3.02 ± 1.62	2.89 ± 1.19	0.450	0.654
Depression	17.09 ± 3.11	16.11 ± 3.76	1.330	0.187
Anxiety	8.59 ± 2.50	8.91 ± 2.37	−0.613	0.542
Fatigue	15.48 ± 3.35	16.18 ± 3.83	−0.918	0.361
Health literacy	42.18 ± 5.90	40.82 ± 6.33	1.045	0.299
Physical activity	4.27 ± 1.48	4.64 ± 1.54	−1.127	0.263
Restrained eating	21.11 ± 8.89	19.48 ± 9.25	0.847	0.400
Emotional eating	18.18 ± 7.37	18.89 ± 7.61	−0.441	0.660
External eating	26.43 ± 9.77	25.61 ± 8.80	0.413	0.681

Pain was assessed by the Numeric Rating Scale (NRS); depression by the Edinburgh Postnatal Depression Scale (EPDS); anxiety by the Generalized Anxiety Disorder Scale (GAD-7); fatigue by the Postpartum Fatigue Scale; health literacy by the Health Literacy Survey–Short Form 12 (HLS-SF12); physical activity by the Pregnancy Physical Activity Questionnaire (PPAQ); eating behaviors (restrained, emotional, external) by the Dutch Eating Behavior Questionnaire (DEBQ).Data are presented as mean ± SD. p-values were calculated using independent t-tests.

At 6 weeks postpartum, baseline-adjusted ANCOVA showed that the intervention group had significantly lower pain, depression, and anxiety scores than the control group, with adjusted mean differences of −0.698 (95% CI: −1.339 to −0.057), −1.313 (95% CI: −2.398 to −0.229), and − 1.357 (95% CI: −2.543 to −0.170), respectively (Table 4). The intervention group also showed higher health literacy and physical activity scores, with adjusted mean differences of 2.714 (95% CI: 0.032 to 5.397) and 0.969 (95% CI: 0.227 to 1.711), respectively. In terms of dietary behavior, emotional eating and external eating were significantly lower in the intervention group after baseline adjustment, whereas restrained eating showed a trend toward a higher score but did not reach statistical significance.

Table 4.

Six-week postpartum outcomes and between-group effect estimates among complete cases (N = 82).

Variable	PCHIP (n = 40)	Control (n = 42)	Estimate (95% CI)	P
Baseline-adjusted outcomes
Pain	1.95 ± 1.01	2.67 ± 1.76	−0.698 (−1.339 to −0.057)	0.033
Depression	17.85 ± 2.69	19.12 ± 2.14	−1.313 (−2.398 to −0.229)	0.018
Anxiety	9.07 ± 2.32	10.43 ± 2.98	−1.357 (−2.543 to −0.170)	0.026
Fatigue	16.15 ± 4.87	17.71 ± 4.39	−1.646 (−3.724 to 0.433)	0.119
Health literacy	42.65 ± 5.54	39.93 ± 6.48	2.714 (0.032 to 5.397)	0.047
Physical activity	3.45 ± 1.99	2.52 ± 1.29	0.969 (0.227 to 1.711)	0.011
Restrained eating	21.12 ± 8.69	17.52 ± 6.98	3.398 (−0.085 to 6.881)	0.056
Emotional eating	17.20 ± 5.23	21.60 ± 10.98	−4.421 (−8.177 to −0.666)	0.022
External eating	25.15 ± 7.98	29.21 ± 9.58	−4.043 (−7.957 to −0.129)	0.043
Weight outcomes adjusted for baseline weight
Postpartum BMI	23.77 ± 3.96	24.72 ± 3.60	−1.025 (−2.531 to 0.481)	0.179
Weight change (kg)	5.78 ± 3.96	3.23 ± 3.74	2.527 (0.835 to 4.218)	0.004
Recovery outcomes without baseline measures
Red lochia duration (weeks)	3.29 ± 1.15	4.51 ± 1.48	−1.224 (−1.811 to −0.638)	< 0.001
Uterine size by ultrasound	0.15 ± 0.36	0.21 ± 0.42	−0.064 (−0.236 to 0.107)	0.458
Resting muscle score	59.42 ± 23.16	48.37 ± 22.14	11.049 (1.095 to 21.003)	0.030
Fast-twitch muscle score	74.44 ± 12.40	71.67 ± 11.80	2.765 (−2.554 to 8.083)	0.304
10-s slow-twitch score	79.60 ± 10.80	72.23 ± 14.73	7.366 (1.666 to 13.066)	0.012
60-s slow-twitch score	76.32 ± 11.82	70.99 ± 12.93	5.328 (−0.126 to 10.782)	0.055
Post-resting score	66.10 ± 21.78	53.93 ± 24.73	12.171 (1.909 to 22.434)	0.021
Breastfeeding status
Formula only	2 (5.0%)	5 (11.9%)
Mixed, mostly formula	3 (7.5%)	6 (14.3%)
Partial breastfeeding	5 (12.5%)	9 (21.4%)
Predominant breastfeeding	10 (25.0%)	11 (26.2%)
Exclusive breastfeeding	20 (50.0%)	11 (26.2%)
Exclusive/predominant	30 (75.0%)	22 (52.4%)	Risk diff. 22.6% (2.4 to 42.8%)	0.041

Data are presented as mean ± SD for continuous variables and n (%) for categorical variables. For outcomes measured at both baseline and 6 weeks postpartum, between-group effects were estimated using ANCOVA with the baseline value of the corresponding outcome as a covariate. Effect estimates are adjusted mean differences, calculated as intervention minus control. For weight-related outcomes, baseline postpartum weight was included as a covariate because weight change was derived from maternal weight measured within the first three days postpartum and maternal weight measured at the 6-week follow-up. For recovery outcomes without corresponding baseline measurements, unadjusted between-group mean differences with 95% confidence intervals are reported. Breastfeeding status was summarized descriptively; exclusive or predominant breastfeeding was compared between groups using Fisher’s exact test. P values should be interpreted cautiously because multiple outcomes were assessed and no formal adjustment for multiple comparisons was applied. Large SDs for several outcomes reflect substantial inter-individual variability in early postpartum recovery and should be considered when interpreting the precision and clinical relevance of the findings.

For weight-related outcomes, estimates were adjusted for baseline postpartum weight. After adjustment, the intervention group still showed greater weight reduction at 6 weeks postpartum than the control group, whereas postpartum BMI at 6 weeks did not differ significantly between groups. For recovery-related outcomes without corresponding baseline measurements, unadjusted between-group estimates were reported. Compared with the control group, the intervention group had a shorter duration of red lochia and favorable pelvic floor muscle findings, including higher resting muscle scores, 10-second slow-twitch muscle scores, and post-resting scores. However, no significant between-group differences were observed for fatigue, uterine size by ultrasound, fast-twitch muscle score, or 60-second slow-twitch muscle score. Exclusive or predominant breastfeeding was more frequent in the intervention group than in the control group.

Discussion

In this trial, the relatively low attrition rate suggests that the digital format was acceptable to most completers, although complete-case analysis limits inference to participants with 6-week follow-up data. The interactive design of PCHIP, including feedback, reminders, and tailored support, may have supported engagement^29,30.

At 6 weeks postpartum, baseline-adjusted analyses showed that women in the intervention group had lower levels of pain, depression, and anxiety than controls. These findings suggest that the program may have helped alleviate selected physical and psychological symptoms during early postpartum recovery^31,32. The benefits may be related to the multidimensional design of the intervention, which integrated dietary guidance, physical activity promotion, interactive communication, and literacy-oriented support³³.

These findings should also be considered within the sociocultural context of China. During zuo yuezi, women are expected to comply with strict behavioral and dietary restrictions, which, while promoting rest, may also intensify psychological stress and role conflict. The rapid transition to motherhood, coupled with infant care responsibilities and limited mobility, can exacerbate anxiety and depression when family support is insufficient. The present intervention mitigated these risks by embedding family-centered support, encouraging shared responsibility in diet and exercise, and providing continuous online communication to reduce isolation. This family-inclusive, interactive approach helped participants navigate role transitions with greater confidence and autonomy, thereby reducing emotional distress.

Participants in the intervention group showed higher health literacy and physical activity scores than the control group after adjustment for baseline values. This finding is consistent with Nutbeam’s model, which conceptualizes health literacy as comprising functional, interactive, and critical skills that enable individuals to access, interpret, and apply health information¹³. In the postpartum period, these skills may support women in translating health information into practical self-care behaviors³⁴.

In line with this interpretation, the intervention group demonstrated lower emotional eating and external eating after baseline adjustment, while restrained eating showed a favorable trend but did not reach statistical significance. These findings suggest that the program may have influenced both emotional and environmental drivers of dietary behavior, although the evidence for restrained eating should be interpreted cautiously³⁵.

Breastfeeding outcomes also favored the PCHIP group. This may be related to timely digital guidance, family involvement, and problem-solving support during the early postpartum period, when breastfeeding confidence and routines are still being established³⁶.

The behavioral improvements observed may also be related to selected maternal recovery outcomes. In unadjusted analyses of outcomes measured only at 6 weeks postpartum, women in the intervention group experienced greater postpartum weight reduction and shorter duration of red lochia than those in the control group. Because these outcomes did not have corresponding baseline measurements, the results should be interpreted as unadjusted between-group differences³⁷.

In the context of zuo yuezi, PCHIP provided a pragmatic way to balance traditional postpartum rest with evidence-based guidance on diet, activity, and recovery³⁸. However, these recovery-related findings were unadjusted and should be confirmed in larger trials.

In this trial, women in the intervention group showed favorable pelvic floor muscle (PFM) performance in selected domains measured at 6 weeks postpartum, particularly resting muscle score, 10-second slow-twitch muscle score, and post-resting score. These findings suggest potential benefits of the structured PFM rehabilitation component, especially the targeted training videos introduced in the third week^39,40.

These PFM findings may reflect the accessible home-based video demonstrations and reminders, which helped participants practice rehabilitation without conflicting with cultural expectations of postpartum confinement⁴¹.

Despite these benefits, no significant between-group differences were observed for fatigue or fast-twitch muscle strength. Fatigue is one of the most common postpartum symptoms and is influenced by multifactorial determinants such as disrupted sleep, breastfeeding demands, and psychosocial stressors—factors not directly addressed by dietary guidance, moderate activity, or PFM training⁴². Notably, although physical activity increased significantly in the intervention group, this did not exacerbate fatigue, highlighting the safety and feasibility of structured, literacy-guided activity interventions during early postpartum recovery.

Similarly, the absence of improvements in fast-twitch muscle strength is consistent with the program’s endurance-focused design. Slow-twitch fibers respond well to moderate, sustained training, whereas fast-twitch adaptations typically require high-intensity or resistance-based protocols⁴³. Given the relatively short intervention period (six weeks) and the safety considerations of early postpartum care, such adaptations were not anticipated.

Taken together, these findings suggest that embedding PFM rehabilitation within a health literacy framework may be feasible and potentially beneficial during early postpartum recovery. At the same time, the absence of significant effects on fatigue, fast-twitch muscle strength, and 60-second slow-twitch performance underscores the need for longer-term and more comprehensive interventions. Future programs might incorporate components such as sleep hygiene education, stress management, and progressive resistance training, while maintaining careful safety monitoring.

Limitations and future implications

Several limitations should be acknowledged when interpreting the findings of this study. First, the relatively short follow-up period of six weeks limits the ability to evaluate long-term effects on maternal weight retention, pelvic floor recovery, breastfeeding continuation, psychological health, and sustained behavior change. Future longitudinal studies extending to 3–6 months or beyond are warranted.

Second, although randomization minimized baseline imbalance, the study was conducted in a single tertiary hospital in eastern China and included a relatively small sample. This limits the generalizability of the findings to women with different socioeconomic, cultural, obstetric, or healthcare-access backgrounds. Multicenter trials with larger and more heterogeneous samples are needed.

Third, the primary analysis was a complete-case analysis of participants with available 6-week follow-up data. Because six randomized participants did not provide post-intervention data, attrition bias cannot be fully excluded. In addition, relatively large SDs were observed for several outcomes in Table 4, suggesting considerable inter-individual variability in early postpartum recovery. Although randomization, blinded assessment, standardized instruments, and baseline-adjusted analyses were used to reduce bias and confounding, residual confounding and imprecision remain possible.

Fourth, multiple outcomes were assessed and no formal adjustment for multiple comparisons was applied. Therefore, statistically significant findings, particularly for secondary and recovery-related outcomes, should be interpreted as exploratory and confirmed in future adequately powered trials. Some behavioral outcomes, including dietary intake, breastfeeding, and physical activity, were partly based on self-reported data and may be subject to recall and social desirability bias.

Finally, while the digital format may improve access for some postpartum women, it may exclude women without stable internet access or sufficient digital literacy. Future iterations could evaluate hybrid models combining online support with in-person coaching, especially for rural or low-resource populations. Despite these limitations, the study provides preliminary evidence that integrating health literacy enhancement with digital behavioral support may be a promising direction for postpartum care.

Conclusion

This randomized controlled trial suggests that PCHIP, a health literacy-driven digital postpartum care model, may improve selected postpartum self-care behaviors and recovery-related outcomes among primiparous women during early postpartum care. Its main conceptual contribution is the integration of health literacy enhancement, behavior change techniques, family-supported practice, and digital feedback into one structured pathway for translating postpartum health information into self-care behavior.

These findings should be interpreted cautiously given the single-center design, small sample size, short follow-up period, complete-case analysis, and multiple outcomes assessed. Larger multicenter trials with longer follow-up are needed to confirm the effectiveness, safety, generalizability, and sustainability of health literacy-based digital interventions for postpartum recovery and self-care.

Author contributions

Conceptualization, Q.L.; methodology, Q.L.; formal analysis, Q.L, and Z.W.; investigation, Q.L, and Z.W.; data curation, Q.L, and Z.W.; writing—original draft preparation, Q.L, and Z.W.; writing—review and editing, Q.L, and Z.W.; project administration, Q.L.; funding acquisition, Q.L. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Social Development—Basic Research Program of the Zhenjiang Municipal Science and Technology Bureau, Jiangsu Province, China (Grant No. FZ2024066).

Data availability

The data presented in this study are available upon request from the corresponding author. The data are not publicly available due to privacy issues.

Declarations

Competing interests

The authors declare no competing interests.

Informed consent

Informed consent was obtained from all subjects involved in the study.

Institutional review board statement

Approval was obtained from the Clinical Research Ethics Committee of the Affiliated Changzhou No. 2 People’s Hospital of Nanjing Medical University, Changzhou, China ([2024]KY311-01, January 10, 2025). The trial was prospectively registered with the Chinese Clinical Trial Registry (ChiCTR2500096473) on January 24, 2025, before enrolment of the first participant.

Declaration of generative AI in scientific writing

Generative AI and AI-assisted technologies should only be used in the writing process to improve the readability and language of the manuscript.