Evaluating the feasibility of Swedish massage as an adjunct approach to care for pregnant women who have experienced a prior stillbirth: a convergent parallel mixed-methods single-arm feasibility study protocol - Statistical Analysis Plan
Women experiencing pregnancy after stillbirth have higher levels of anxiety, fear and depression than women who have no history of loss. Qualitative studies show that standard antenatal care may be emotionally unsuitable for many women at this time. However, there is a lack of evidence on what interventions or approaches to care might benefit these women. Therapeutic massage may assist women after stillbirth by decreasing anxiety, worry and stress.
This paper outlines the statistical analysis plan for the Helping suppOrt individuals Pregnant after Experiencing a Stillbirth (HOPES) feasibility trial which evaluates massage as an adjunct approach to care for pregnant women who have experienced a prior stillbirth.
HOPES is a convergent parallel mixed-methods, single-arm repeated measures trial conducted in trained massage therapists' private clinics across Australia. HOPES aims to recruit 75 individuals pregnant after a previous stillbirth. The intervention is massage therapy treatments, and participants will receive up to five massages within a 4-month period at intervals of their choosing.
Primary quantitative outcomes are the feasibility and acceptability of the massage intervention. Secondary outcomes include determining the optimal timing of massage therapy delivery and the collection of measures for anxiety, worry, stress and self-management. Progress through the feasibility study from approach through to completion of the study will be summarised using a CONSORT statement and flow diagram. The participant’s demographic and clinical characteristics will be summarised using descriptive statistics. Summary statistics of the observational clinical treatment data (e.g., worry, anxiety, etc.) will be reported and an appropriate matched-test will be used to determine significant differences between baseline and post final treatment scores (e.g. paired t-test if data are normally-distributed, Wilcoxon matched-pairs test if not).