Premenstrual syndrome (PMS) and premenstrual dysphoric disorder (PMDD) affect a significant portion of women, leading to reduced well-being and even increased suicidality. Given the high smartphone ownership rates and the demonstrated efficacy of mental health apps in general, there's a clear opportunity to develop effective apps specifically targeting PMS and PMDD symptoms. However, currently, few such apps exist, despite the potential for convenient and accessible support. This study aims to fill a knowledge gap by exploring user preferences for key app features and identifying the health behavior factors that influence the intention to use such apps. The study uses the Health Belief Model (HBM) as a theoretical framework to understand how perceived benefits, barriers, severity, susceptibility, self-efficacy, and cues to action influence the intent to utilize a premenstrual mental health app. The importance of this research lies in its potential to guide the design and development of more effective and engaging mental health apps for women experiencing PMS and PMDD, ultimately improving their well-being and access to support.

Existing literature highlights the prevalence and impact of PMS and PMDD on women's lives, emphasizing the need for effective interventions. Studies show the general acceptability and efficacy of mental health apps for various conditions, suggesting a promising avenue for addressing premenstrual mental health symptoms. While some period tracking apps incorporate mood tracking, dedicated apps for PMS and PMDD are limited. There is evidence supporting the efficacy of internet-delivered cognitive behavioral therapy (CBT) and online peer support for these conditions, suggesting potential for app-based delivery. The Health Belief Model (HBM), a well-established framework for understanding health behaviors, has been successfully applied to various health interventions, including mobile health interventions and mental health help-seeking behavior, making it suitable for this study. However, limited research exists specifically on user preferences and factors influencing the use of apps for premenstrual mental health symptoms, justifying the present study.

This study employed an online survey distributed through various channels (email, word of mouth, paid advertisements, and social media) between January and March 2023. The participants (N=530) were women aged 18 and over residing in the UK, who had experienced premenstrual mental health concerns, were not pregnant, breastfeeding, or in perimenopause/menopause. The survey assessed sociodemographic characteristics, premenstrual symptoms using the Premenstrual Symptom Screening Tool (PSST), preferences for app features, willingness to pay for an app, and health beliefs using a 25-item questionnaire based on the HBM, scored on a 6-point Likert scale. Data analysis included descriptive statistics, Cronbach's alpha for internal consistency, Spearman rank-order correlations, Mann-Whitney U tests, and structural equation modeling (SEM) using Stata 17.0 to assess the HBM's predictive power on intention to use the app. The SEM involved confirmatory factor analysis (CFA) to evaluate model fit, using Satorra-Bentler scaled chi-squared test, Comparative Fit Index (CFI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR) as fit indices.

The study found that symptom monitoring (74.72%) and psychoeducation (57.92%) were the most preferred app features. A significant majority (52.64%) were unwilling to pay for the app; those willing preferred a one-off payment. The SEM revealed a good model fit (χ²(254) = 565.91, p < 0.001; CFI = 0.939, RMSEA = 0.048, SRMR = 0.058), with the HBM constructs explaining 58.22% of the variance in intention to use. Cues to action (β = 0.49, p < 0.001) was the strongest predictor, followed by perceived barriers (β = −0.22, p < 0.001), perceived severity (β = 0.16, p = 0.012), and perceived benefits (β = 0.10, p = 0.035). Self-efficacy and perceived susceptibility were not significant predictors. Analysis of sociodemographic factors revealed that higher educational attainment and annual household income were positively correlated with self-efficacy but negatively associated with perceived severity and intention to use. Employment status also influenced perceived susceptibility and severity.

The findings highlight the importance of user-centered design in the development of premenstrual mental health apps. The strong preference for symptom monitoring and psychoeducation underscores the need to incorporate these features, providing users with tools for self-management and increased understanding of their condition. The significant role of cues to action emphasizes the value of securing healthcare professional endorsement and promoting the app through trusted channels. Addressing perceived barriers, such as digital discomfort, privacy concerns, and perceived app quality, is crucial for improving app uptake. The importance of perceived benefits and severity suggests that clearly communicating the app's potential advantages and acknowledging the seriousness of premenstrual symptoms are necessary to motivate adoption. The lack of significance of self-efficacy and perceived susceptibility in this sample, likely due to the high digital literacy and prevalence of symptoms, suggests that different factors may be at play compared to other populations.

This study provides valuable insights into user preferences and factors influencing intention to use apps for premenstrual mental health symptoms. Key recommendations for app developers include co-designing with users and healthcare professionals, prioritizing free or low-cost options, emphasizing symptom monitoring and psychoeducation, addressing privacy concerns, and clearly communicating the app's benefits. Future research could investigate factors beyond the HBM, explore the translation of intention into actual use, and examine the effects of app usage on mental health outcomes in a more diverse sample.

The study's sample was highly educated and affluent, potentially limiting the generalizability of the findings. The online recruitment method might have biased the sample toward individuals with higher digital literacy and more severe symptoms. The HBM items were not previously validated, potentially introducing measurement error. Finally, the study focused on intention to use rather than actual usage, necessitating further research on sustained engagement.