Background The intervention is informed by the transactional model of stress and cognitive appraisal and the findings of the research programme. It combines perinatal loss health education, ACT-based psychological support, rehabilitation activities including mindful yoga, and family social support. ACT addresses acceptance, cognitive defusion, contact with the present moment, self-as-context, values clarification, and committed action.
Before pilot testing, intervention content was reviewed through two to three rounds of Delphi consultation with eight experts: two psychology experts, three obstetricians, and three obstetric nurses. Each expert had more than 10 years of experience and a senior professional title. Relevance, clarity, readability, and appropriateness were rated on a 4-point scale. Two women with previous perinatal loss and education no higher than high school also reviewed comprehensibility and relevance to their needs. All final expert item ratings were at least 3, and minor revisions informed the pilot intervention.
Intervention This study tests a 4-week ACT-based psychosocial intervention delivered across six sessions (45-60 minutes each), plus a 30-minute booster session one month after the final session. The intervention integrates four thematic components: (1) perinatal loss health education; (2) ACT-based grief and emotion management (acceptance, defusion, present-moment awareness, values, committed action); (3) postpartum rehabilitation mindfulness yoga; and (4) social support from significant others.
Delivery format is mixed to accommodate the inpatient-to-outpatient transition:
Sessions 1-4: In-person, hospital setting, one-on-two format (woman plus spouse/significant other), covering the inpatient period (admission, induction procedure, postoperative recovery).
Sessions 5-6: Remote via Tencent Meeting (videoconferencing), one-on-one format, covering the post-discharge physical and emotional recovery period.
Booster session: 30-minute remote session one month after session 6. Participants receive a workbook containing session summaries and homework assignments. QR codes in the workbook provide access to online instructional videos for home-based mindfulness meditation or mindfulness yoga practice.
Study Design This is a two-arm, assessor-masked, parallel-group pilot randomised controlled trial conducted at three tertiary hospitals in Changsha, Hunan, China. Woman-spouse dyads are allocated in a 1:1 ratio to usual care plus the family psychosocial intervention or usual care alone. Block randomisation uses a block size of four. An independent research assistant, a registered nurse with a bachelor's degree in nursing, generates the allocation sequence using an online randomisation programme and is not involved in recruitment, intervention delivery, or outcome analysis.
Allocation is concealed using sequentially numbered, sealed, opaque envelopes prepared by the research assistant and delivered to participants by the project manager. Outcome assessment uses online questionnaires, with participant identifiers and allocation coded before analysis to support assessor masking. Process evaluation and a descriptive qualitative component inform refinement of the intervention.
Participants, Sample Size, and Recruitment The target sample is 44 woman-spouse dyads, with 22 dyads per arm. The sample-size rationale uses a pilot-study minimum of 30 and an anticipated attrition rate of 24.31% from the preceding systematic review and meta-analysis. This is a pilot sample intended to evaluate feasibility, acceptability, and preliminary effects.
Women aged at least 18 years who require induction of labour because of fetal or maternal health problems are recruited with a spouse willing to participate as the primary caregiver. Participants must be able to use a smartphone, tablet, or computer for online intervention and questionnaires and provide informed consent. Exclusions include life-threatening complications, diagnosed psychological or psychiatric illness or evident self-harm/suicidal tendencies or behaviour, and current or recent participation in a related support intervention within the preceding 30 days.
Convenience sampling is used. After ethics approval at each hospital, midwives distribute recruitment information through hospital online platforms and posters. Interested women undergo eligibility screening through the responsible midwife and meet a research assistant to discuss the study purpose, procedures, benefits, risks, and the right to withdraw before providing written informed consent.
Outcome Measures
Quantitative measures:
The primary preliminary-effectiveness outcome is perinatal grief, measured using the Perinatal Grief Scale (PGS). Secondary outcomes for women and their spouses are depressive symptoms (PHQ-9), anxiety symptoms (GAD-7), post-traumatic stress symptoms (IES-R), psychological flexibility (Psy-Flex-C), health information literacy (HILS), mindful self-care (B-MSCS), and perceived social support (PSSS).
Feasibility is assessed through eligibility and consent rates, reasons for declining participation, adherence, intervention implementation, attrition, and reasons for withdrawal. Intervention completion is defined as attending at least six of the eight sessions (at least 75%).
Acceptability is assessed in the intervention group using an 18-item study-specific satisfaction questionnaire adapted from the 15-item Chinese Standardized Client Satisfaction Scale and qualitative interviews at T1. Each woman and her spouse complete the satisfaction questionnaire jointly. Items use a 5-point Likert response scale from 1 (very dissatisfied) to 5 (very satisfied); higher scores indicate greater satisfaction.
Qualitative component:
All women and spouses in the intervention group (target: 22 dyads) are invited to semi-structured interviews within 1 week after intervention completion (T1). Trained research assistants conduct the interviews through Tencent Meeting to explore experiences, perceived facilitators, barriers, and benefits. Interviews are digitally audio-recorded with consent using an iFlytek SR701 recorder.
The first author transcribes the recordings verbatim and a research assistant checks the transcripts. Content analysis is conducted with NVivo following Bengtsson's four stages: decontextualisation, recontextualisation, categorisation, and compilation.
Assessment Schedule Baseline data (T0), including demographics and primary and secondary outcomes, are collected online with research-assistant support before randomisation. Post-intervention assessment (T1) occurs within 1 week after completion of the approximately 3-month intervention, including the final booster session at 3 months after discharge. The intervention group completes primary and secondary outcome questionnaires and the satisfaction questionnaire and is invited to qualitative interviews during this week. The control group completes only the primary and secondary outcome questionnaires over the corresponding time interval.
Data Management and Quality Control Data are collected through Questionnaire Star (WJX). Records use study identification codes and are password-protected. Paper records are stored in locked cabinets; electronic records are stored on password-protected USB drives with password-protected cloud backup accessible only to the research team. Records are retained for 6 years and then destroyed. Audio recording requires participant consent.
Participation is voluntary, written informed consent is obtained, and participants may pause or withdraw without penalty. If self-harm or suicidal tendencies are reported or identified, the research assistant informs the principal investigator, who facilitates access to medical support and, with participant consent, informs the emergency contact. Intervention contacts include physical and psychological risk review and referral as required.
Data Analysis Quantitative data are analysed using R with a significance level of 0.05. Descriptive statistics summarise demographic and clinical characteristics, implementation measures, and outcome scores. Distributional characteristics are examined using skewness, kurtosis, normality tests, histograms, and Q-Q plots. Baseline continuous variables are compared using independent-samples t tests or Mann-Whitney U tests as appropriate; categorical variables are compared using continuity-corrected chi-square tests or Fisher's exact tests.
For non-normally distributed outcome data, within-group changes from T0 to T1 are examined using Wilcoxon signed-rank tests and between-group comparisons use Mann-Whitney U tests. Between-group effect sizes at T1 are estimated using Hedges' g.
Qualitative data are analysed with NVivo using Bengtsson's content-analysis stages of decontextualisation, recontextualisation, categorisation, and compilation. Recordings are transcribed verbatim by the first author and checked by a research assistant.
Study Objectives The pilot evaluates whether the family psychosocial intervention can be delivered as planned, its acceptability to women and their spouses, and preliminary differences between the intervention and usual-care groups in changes in perinatal grief and secondary outcomes from T0 to T1. Feasibility and acceptability findings will inform intervention refinement. Intervention completion is defined as attendance at six or more of the eight sessions.
