A mixed-methods study to evaluate the effectiveness and cost-effectiveness of aerobic exercise for primary...

Several studies have evaluated the effects of high-intensity aerobic training (HIAT) on pain severity and quality of life (QoL) among women with primary dysmenorrhea. However, to date, no studies have evaluated the effectiveness of HIAT on academic performance or absenteeism or examined the cost-effectiveness of HIAT relative to other treatments in women with primary dysmenorrhea. Furthermore, the mechanisms underlying aerobic exercise-induced analgesia in primary dysmenorrhea remain unclear. The objectives of this study are to: (1) evaluate the effects of HIAT on absenteeism and academic performance among university students, (2) identify the underlying mechanisms associated with aerobic exercise-induced analgesia in primary dysmenorrhea, and (3) determine the cost-effectiveness of HIAT compared with a wait-list control (WLC) group receiving usual care.

A sequential, embedded, mixed-methods study design, including a crossover, randomised controlled trial (RCT) and semi-structured focus groups, will be conducted alongside an economic evaluation. A total of 130 women aged 18-24 years will be randomised into either HIAT (n = 65) or wait-list control (n = 65) groups. Primary outcomes will include average pain intensity, absenteeism from university, and academic performance. Primary mediators will include salivary progesterone and prostaglandin F2α levels. Outcome and meditator variables will be assessed at baseline and post-treatment, at 12 and 28 weeks. An economic analysis will be conducted from the societal and healthcare perspective of Hong Kong. Semi-structured focus groups will be conducted at 32 weeks. Of the 130 participants included in the RCT, 70 will be included in the focus groups.

All statistical analyses will be performed on an intention-to-treat basis, using SPSS (version 24). Preliminary analysis using an independent samples t-test and a two-sided, unpaired Student's t-test will be performed to exclude carryover effects and identify within-participant differences in outcome variables between the study periods, respectively. Treatment effects will be evaluated using analysis of variance via a mixed-effects model with fixed effects for intervention, period, and sequence. In all models, random effects will include the participants nested within the sequence as a sampling cluster. The mediation effects will be assessed using the Sobel test. The EQ-5D responses will be converted into utility scores to estimate the gain or loss of quality-adjusted life-years. Seemingly unrelated regression analyses will be used to estimate the total cost differences and effect differences. Qualitative data will be analysed using the process of thematic analysis.