Evaluation of a Web-Based Resource to Improve Menstrual Health Literacy and Self- Management in Young Women

Menstrual cycle-related conditions, such as dysmenorrhea and heavy bleeding, are common amongst those under 25. Despite having signicant impact on work, education, and social activities, most do not seek medical advice, preferring to self-manage their symptoms. We aimed to determine if access to a web-based resource could improve health literacy, reduce menstrual symptoms, and encourage health-seeking behaviour in those at risk of conditions such as endometriosis. People were eligible to participate if they were currently living in Australia, aged 14-25 yrs and had menstruated for at least 12 months. Access to the resource was for three menstrual cycles. Seventy-ve participants with a mean age of 20.4 years enrolled to use the web-based resource. Forty-three percent visited their doctor regarding their menstrual symptoms during the study period, with 84% indicating that they made the appointment due to the resource; over half (56%) who visited their doctor received a referral to a gynaecologist. There were clinically relevant improvements in health literacy reecting improved condence of participants regarding having adequate information and skills to manage their symptoms. A low-cost, web-based resource can improve health literacy, provide effective options for self-management of menstrual cycle-related conditions, and motivate young people to see medical advice.


Introduction
Menstrual conditions, such as dysmenorrhea (period pain) and heavy menstrual bleeding (HMB), are common in young women worldwide 1,2 . Dysmenorrhea is very common in young Australian women, with research showing over 90% of those under 25 reporting regular dysmenorrhea 3,4 . Despite the signi cant impact menstrual symptoms such as dysmenorrhea have on academic attendance and performance 3-5 , many women, regardless of age, think that pain is a 'normal' part of being a woman 6 . Most of these women do not seek medical help but rather use self-management 7 . However, they are often unsure which self-management strategies are effective and often report only partial symptom relief 7,8 .
The Institute of Medicine de nes health literacy as "the degree to which individuals have the capacity to obtain, process, and understand basic health-related decisions" 9 . In an educational context, the Australian Health and Physical Education (HPE) curriculum 10 focuses on developing knowledge, understanding and skills related to the three dimensions of health literacy: functional, interactive and critical dimensions, where health literacy is considered a personal and community asset to be developed, evaluated, enriched and communicated. With a speci c focus on menstrual health literacy, the HPE curriculum includes sexuality and sexual and reproductive health, puberty, and access to health services.
However, almost 50% of Australian teachers report they are under-prepared to discuss menstruation as part of the normal curriculum 11 . This unpreparedness from teachers may explain why young women and their parents in Australia consistently report low menstrual health literacy 7,12,13 and why young women often rate their period as 'normal' despite signi cant related absenteeism and regular non-cyclical pelvic pain symptoms 7 . This is especially concerning when combined with the low rates of seeking medical attention, as almost half of young women cannot identify the signs of secondary dysmenorrhea such as endometriosis 7 . A signi cant part of this issue is that menstruation is often explained only insofar as women understand enough to protect themselves from unwanted pregnancy, rather than empowering them to understand how it affects their lives 14 .
Increasing health literacy through menstrual education can assist women's understanding, positively affecting their ability to engage in self-care and help identify symptoms that require further investigation 15,16 . In addition, there are a number of effective, non-pharmacological interventions that can reduce menstrual symptoms including heat, low intensity exercise, such as yoga or stretching, and acupressure [17][18][19][20] , however most young women appear unaware of these 8 . Strategies to raise awareness of these interventions are thus required.
This study aimed to evaluate a web-based resource delivering menstrual education, self-screening for menstrual disturbance and evidence-based self-care, to determine if it was acceptable and feasible for young people and if this resource could increase their menstrual health literacy, improve engagement with health professionals, change their management strategies and reduce their menstrual pain.

Study design:
Ethical approval for this study was provided by the Western Sydney University Human Ethics Committee (H12599 -March 2018). All research complied with the relevant guidelines and regulations outlined in the National Statement on Ethical Conduct in Human Research (2018) 21 . All participants (and their legal guardians if aged <18yrs) provided written informed consent prior to the start of the study. The webbased resource was hosted by Western Sydney University and was developed based on the results of our national survey on menstrual health literacy 7 , as well as our team's previous work on exploring effective non-pharmacological self-management strategies for dysmenorrhea and other menstrual symptoms such as Pre Menstrual Syndrome (PMS) [17][18][19] . The resource was co-created with an expert steering group consisting of the research team, two gynaecologists, two general practitioners (GPs), two secondary HPE teachers, a sexual and reproductive health curriculum expert, and ve young women (16-25yrs) from diverse backgrounds. The steering group met twice, rst to discuss the content and format of the information that should be included in the resource and again approximately four weeks later, to assess the suitability of the resource content developed by the research team. Minor changes were made in response to the feedback from the steering group. The resource was developed with two main sections: Section one was evidence-based information on the menstrual cycle, covering hormonal changes during the menstrual cycle, the signs, symptoms and suspected causes of dysmenorrhea (both primary and secondary), PMS and heavy menstrual bleeding. An emphasis was placed on the range of symptoms that could be considered 'normal' and those that were likely to require further medical investigation, encouraging participants to speak to their doctor about these. The second section contained immediate self-care options including guidance on how to use heat effectively, and information on the usage of over-the-counter medications, such as ibuprofen, and a range of medium to long term self-care options including, acupressure points, a yoga sequence (Yoga Nidra) and dietary guidance. On the advice of the steering group, self-management information was delivered as a mixture of text, instructional video clips and audio recordings. Once the resource was developed, it was pilot-tested with four young women under the age of 21, and further editorial changes were made to the content to ensure it was clear and understandable for this age group.
Participants, recruitment and study ow: People were eligible to participate if they were currently living in Australia, aged 14-25 yrs, had menstruated for at least 12 months and had regular periods (within 21 to 35 days), and had internet access on their computer, tablet or smartphone.
Participants were recruited using a variety of methods. These included paid Facebook and Instagram advertising and social media promotion by Family Planning NSW, Endometriosis Australia, ACHPER and Kimberly-Clark Australia as well as via the social media networks (both personal and professional) of the research team. Potential participants lodged their interest via a web-portal and a member of the research team (K.P.) contacted them via email. A phone call was then undertaken (if under 18 including the parents/guardians) to get consent and screen participants.
Participants then completed baseline measures after their next menstrual period and subsequently provided with web-resource login and password. The Period ImPact and Pain Assessment (PIPPA) tool was previously developed to assist young women to determine normal from atypical period pain and interference 4,22 . At baseline, all participants lled in an online version of the PIPPA tool, which asked ve Y/N questions regarding the impact of their menstrual symptoms. If they answered "Yes" to either four or ve of the PIPPA questions, they received an email with a personalized letter advising them to make an appointment and take this letter to their doctor. The letter contained an explanation of the PIPPA tool and the related questions they had answered "Yes" to and requested that further investigations be made if the doctor felt it was appropriate. Access to the resource covered three menstrual cycles. They were asked to log in to the resource at least twice in the rst month to read the information section and familiarize themselves with the self-management options, and at least once per month for the next two months. Participants were encouraged to use as many or as few self-management techniques as they preferred.
Regular automated reminders, either via email or SMS/text messages (based on participant preference), were sent twice per month to remind participants to use their self-management method(s) of choice. The total duration was approximately four months, with three menstrual cycles using the resource with a onemonth follow-up period.
Sample size: As this was an evaluation study with a single arm no formal sample size calculation was undertaken. Given the lack of previous data on the effectiveness of such web-based interventions, a decision was made to include 75 participants to ensure that rare events and/or small effect sizes would be captured 23 .
Outcome measures and data analysis: Table 1 outlines the outcome measures and when these were collected. All data was captured using Castor EDC 24 .
Descriptive statistics, including frequencies and percentages, means and standard deviations, or median and interquartile range (IQR), were used to report recruitment and dropout rates and demographics information. All outcomes were determined a priori, and as this was a pilot study, all p-values are considered exploratory. All parametric outcomes were compared using paired t-tests, while nonparametric outcomes such as PIPPA scores were compared using the Wilcoxon signed-rank test. Missing data were dealt with via last observation carried forward (LOCF) on the assumption that those who dropped out of the trial were non-responders.

Results
Two hundred and seventy-four young people expressed interest in using the resource over a three-month recruitment period. The most common reasons for exclusion were lack of response to screening phone calls (n=124, 45%) and being outside the age range for eligibility (n=35, 13%). Seventy-ve (27%) were eligible to participate, consented and entered the study, with 56 (75%) using the resource for the full three menstrual cycles. The most common reasons for non-completion were lack of time (n=9, 47%) or were lost to follow-up (n=6, 32%) (de ned as no response after 3 weeks of attempted contact), pregnancy (n=2, 11%) or menstrual suppression due to contraceptive use (n=2, 11%). Table 2 outlines the demographics and menstrual characteristics of the participants. Participants had a mean age of 20 (3.4) years, and the majority were currently at school or university (60%). Eighteen participants (24%) had a con rmed diagnosis of endometriosis. Almost three quarters (73%) of the participants had previously seen a doctor for menstrual symptoms. However, only 16% were happy with the advice they received from their health professional. Eighty per cent of participants were at or had attended co-educational schools, with just over half currently attending or had attended a public school (52%), while under a quarter (22.5%) attended a religiously a liated school or a private school (17.5%). Only 40% reported they had learned about menstrual health in their Health and Physical Education (HPE) classes.
Those with a diagnosis of endometriosis (n=18) had a median PIPPA score of 5, while those without (n=57) had a signi cantly lower median score of 3 (P=.04). Thirty-eight participants (51%) had a PIPPA score of 4 or greater and received a referral letter for their GP. Half of those who received the referral letter subsequently made an appointment with their GP.
Just under half (32 participants, 43%) of the study participants visited their doctor regarding their menstrual symptoms during the study period. Of those who visited their doctor, 27 participants (84%) indicated that they made the appointment due to being involved in the study; either due to information they received on the website (8 participants, 30%) or the referral letter they received (19 participants, 70%). Out of the 32 participants who visited their GP, over half (56%) had received a referral to a gynaecologist by the end of the one month follow up period.
Menstrual management: Table 3 outlines the changes in speci c self-management strategies that were part of the study between baseline and the third menstrual period. Of that 84% of participants using analgesics, the most common analgesics used prior to the study were Ibuprofen (33%), Paracetamol (33%) and Mefenamic acid (18%). There were no signi cant changes (P=.67) in the type of analgesic used or the dosage used (P=.72) between baseline and the third menstrual period. Overall, use of self-management for menstrual symptoms increased from 65% of participants at baseline to 86% at the third menstrual period.
At the end of the one-month follow-up, 60% of participants reported that the resource changed the way they managed their menstrual symptoms, and 85% would keep using at least one of the self-care practices they had learned.

Menstrual pain and symptoms:
Those with a diagnosis of endometriosis (n=18) had a signi cantly (P=.005) higher menstrual pain score at baseline (7.7) compared to those without (n=57) a diagnosis of endometriosis (5.6). Overall, there was a small, non-signi cant (P=.22) decrease in menstrual pain from 6.1 (1.9) at baseline to 5.3 (2.2) at the end of the three months. At one-month follow up, 4% said their menstrual symptoms were 'very much better', 1 in 5 (20%) said their menstrual symptoms were 'much better' than before they started, almost half (46%) said 'a little better', and just over a quarter (28.6%) said no change.

Internal Health Locus of Control
There was a small, non-signi cant (P=.24) increase in mean internal Health Locus of Control from 29.19 (3.39) at baseline to 29.96 (3.25) at one-month follow up.
Health literacy measures:

Menstrual health literacy:
Prior to beginning the study at baseline, only a quarter of participants (25%) felt that they had su cient information about their period and its symptoms, with just over half (53%) reporting that they had some information but felt unsure about a number of aspects, with 15% being unsure if they had enough information at all. At the end of the follow-up period, 59% of participants indicated that they felt they had su cient information about their period and its symptoms, 35% reported that they had some information but felt unsure about a number of aspects, with 6% being unsure if they had enough information at all.
At baseline, when asked about speci c aspects of menstrual health, 60% indicated they knew the signs of heavy menstrual bleeding, and 69% could identify the symptoms of endometriosis. At follow-up, 87% could identify the signs of heavy menstrual bleeding, and 94% could identify the symptoms of endometriosis.
At the end of the follow-up, just under half of the participants (48.8%) said the resource had changed what they thought was a 'normal' period and over a third (35%) said it corrected previous misunderstandings they had about periods and the menstrual cycle. After using the resource, 65% said they were now 'very comfortable' talking about periods to others.
Participant satisfaction: 85% found the web-based resource easy to use, and 90% found the information provided 'very helpful'. A quarter of participants (25%) were 'very satis ed' with the changes in their symptoms, and over a third (34%) were moderately satis ed. Only one participant was unsatis ed.

Discussion
Our study found that access to an online resource for three menstrual cycles could improve some key aspects of menstrual health literacy, encourage engagement with medical professionals, and provide improved understanding of both a 'normal' menstrual cycle and how to identify problematic menstrual conditions such as endometriosis and heavy menstrual bleeding.
There was an increased uptake of self-care measures over the course of the study, especially in breathing/meditation exercises and acupressure which were low at baseline. There is some preliminary evidence that mindfulness style interventions reduce menstrual pain 25,26 and there is evidence for both mindfulness and acupressure for reducing PMS symptoms 19,27 . Exercise has the most signi cant reduction in pain compared to other self-care such as heat, 17 but uptake was lower than other forms of self-care. This may be due to the fact that our resource aimed to increase health literacy, but content was not included to support behaviour change. Incorporating behaviour change theory-based content such as stepping individuals through use of behaviour change techniques 28,29 , e.g., using self-monitoring, setting goals, social support, and/or adding behaviour change support intervention methods in addition to the online platform, e.g., face-to-face methods, could further enhance the e cacy of the resource in regards to changing behaviours of seeking health care and using self-management strategies, such as exercise. Our team are currently working to develop such additional support, including the testing of a menstrual pain behaviour app 30 . Despite the increased uptake of these self-care measures, we did not nd evidence of a clinically signi cant reduction in menstrual pain, however it is important to note that this study was not powered to detect any such change if it did exist. The pain score itself is also unlikely to capture the complete spectrum of bothersome menstrual symptoms such as emotional changes 6 , which may explain why, despite the lack of a signi cant pain reduction, over half of the participants rated their menstrual symptoms as at least slightly improved.
The addition of a menstrual disturbance self-screening tool (PIPPA) to menstrual education in this study demonstrated health seeking behaviour by women to GPs and reported action by GPs to refer some participants to gynaecologists. Young women commonly report GPs being dismissive of pain and interference related menstrual disturbance 31,32 and this can cause reluctance to engage with health-care professionals 6 . In this study, the addition of a letter to GPs with PIPPA scores and explanation of relevance may well have provided a reverse education of GPs to take seriously young women's reports of menstrual disturbance. The action by GPs to refer women to gynaecologists may well have been facilitated by the combination of a high PIPPA score, providing external validation of pain and interference, and women's increased menstrual health literacy and con dence to present to GPs. Given that over half of young Australian women report symptoms of non-cyclical pelvic pain but also normalise these symptoms 3 , ensuring that these young women are advised and empowered to seek medical advice is vital to ensure that the lengthy diagnostic delays are reduced 33,34 .
Certain aspects of health literacy related to agency, empowerment and improved health management skills all increased during the study period. The signi cant improvement in the HSI domain of the HLQ corresponds to "Feel(ing) con dent that they have all the information that they need to live with and manage their condition and to make decisions" 35 . Several domains of the HeiQ also showed reliable improvement, with the largest improvement in the skills and technique acquisition domain; improvement in this domain relates to "highly developed skills in symptom relief and techniques to manage own health" 36 . The proportion of reliable positive change in our participants was similar across most domains to previous research on the impact of chronic disease management programs in Australia between 2007-2013 37 suggesting that, despite the online nature and minimal interaction between the research team and participants, or between participants themselves, it was similarly effective to more 'traditional' programs. Interestingly, despite these improvements there was no corresponding improvement in the internal Health Locus of Control scores over the course of the trial, but the IHLoC scores both at baseline and the end of study were similar to previous data on those with and without a chronic health condition 38 , suggesting that these scores were already in the normal range.
These ndings illustrate that a relatively large number of young people are; not receiving accurate information in schools, from parents or healthcare professionals, information is not being delivered in a suitable manner, or they are missing the information they need. Previous research in Australia has shown that when discussing a condition such as endometriosis young women wanted to know speci c symptoms, rather than vague or general information 39 . Focussing on high schools, the HPE syllabus does not directly address menstrual health education. Content covered in the later years is linked to sexual and reproductive health through the focus area 'Relationships and Sexuality', which aims to explore resources and strategies to manage personal, physical and social changes and transitions associated with puberty as they grow older 10 . In the later years (years 9 and 10), the focus of this content shifts to sexuality and sexual health education, leaving it to the teacher's discretion to incorporate this content at all or in an effective manner beyond personal experience or from a biological lens. Given the signi cant number of participants attending co-educational schools (80%), the ndings of this study reveal gaps in the HPE curriculum between the menstrual health education delivered in school and what information is needed to develop young people's menstrual health literacy.
We recommend a whole-school approach to develop menstrual health literacy and strategies to support students, teachers and school staff to promote menstrual health and wellbeing. This research indicates the need for greater emphasis on menstrual health in HPE classes. Menstrual health education needs to focus on correcting mistaken assumptions about a 'normal' period or menstrual cycle, providing su cient information about periods and their symptoms, and information about menstrual management, including self-care options and when to consult a doctor. To develop young peoples' menstrual health literacy within the critical dimension, the HPE curriculum needs to provide menstrual health education that empowers students to selectively access and critically analyse menstrual health information and to take action to promote their menstrual health 10 . Such information could potentially help to alleviate not only dysmenorrhea and related menstrual symptoms but could, by extension, increase their ability to engage fully in the classroom and extra-curricular activities. The provision of an online resource, such as the one reported in this paper, clearly offers young people and teachers additional information that they could access autonomously. Providing such a resource would ensure critical information was more broadly disseminated and independent of individual teacher capacity. Furthermore, such a tool provides young women with the knowledge to take charge of their health. Ideally, the use of such a resource would be accompanied by professional learning for teachers and potentially parents and carers of young women, to ensure that the health literacy messages received by young women are aligned.
Translational capacity of interventions is essential to consider when designing and evaluating health interventions, to ensure follow-through to sustainable policy and practice change from research settings.
The Reach, Effectiveness, Adoption, Implementation, Maintenance (RE-AIM) framework 40 , is one of the most commonly used implementation science frameworks. Regarding the current study, reach into the intended target population (e.g., the total number of eligible women that were exposed to the intervention invite), adoption of the intervention (e.g., the number of women out of the total potential eligible sample who joined the study) and implementation of the intervention (e.g., how the intervention was used, such as number of hours of engagement and with which elements of the online resource) were not determined. A range of social media sources were used to recruit women making it challenging to identify the total potential sample and the online platform used for resource delivery did not capture analytics to inform use and engagement with the resource. This study was also a short evaluation study and so long-term engagement with the resource (maintenance) was not explored. Our evaluation tools did however detect meaningful changes in some outcomes, suggesting that further testing is warranted to explore full scale, long-term effectiveness and implementation processes in a rigorous trial. Future work should also collect in-depth qualitative data from resource users (e.g., women) and stakeholders (e.g., teachers) to explore how and why outcomes changed the way they did over time 41 We have suggested that high schools may be an appropriate place to reach young women. However, consideration of other settings to promote the resource in women and important in uencers of women beyond high school should also be considered (such as through universities, pharmacies, healthcare settings, community groups and other social marketing strategies). We did not collect data on ethnicity, educational level or socio-economic status, which are associated with health disparities, potentially through health literacy 42,43 . It may be that our resources would require tailoring to particular groups, which will need explored in future. For example, peer support (support from someone with similar characteristics to the individual and with experiential knowledge) has been shown to be particularly effective for lifestyle behaviour change in culturally diverse communities 44 and so expanding on our current resource to add an element of peer support may be necessary to engage well with such communities.
There are several strengths to this study including the use of validated health literacy questionnaires, and the previously validated PIPPA tool to encourage engagement with health professionals. However, some important limitations must be acknowledged. Firstly, this is a pilot study and therefore was not powered to detect small, even if clinically relevant, changes between baseline to end of trial. Secondly, due to budget constraints and internal University policy, the web-based resource had to be hosted on the universities own web platform. This platform lacks many of the features of the "modern" Web 2.0, and therefore data on which pages were interacted with and for how long, what video clips received the most views etc was not able to be retrieved. Therefore, we are unable to determine how much actual engagement with the material there was. Future versions of this content will be hosted on modern platforms that will allow the research team to use analytics to help optimise content. Finally, the outcomes of the visit to the general practitioner or gynaecologist are unknown, therefore we are unsure about how many of the participants, if any, received a further diagnosis such as endometriosis for their symptoms.
A web based resource, with minimal setup and ongoing costs, can provide clinically relevant improvements in health literacy, remedy previous misconceptions about menstruation and increase awareness and up-take of evidence based self-care methods. The used PIPPA tool was a simple but effective means of empowering and encouraging visits to health professionals. Future research should explore the most acceptable way to deliver this information, and to ensure that not only young people, but their teachers and parents/guardians also have access to accurate, evidence based information on menstruation and it's management. The Health Education Impact Questionnaire (HEIQ) is a perception-and evaluation-based measure that was developed to be a user-friendly, relevant and psychometrically sound instrument for the comprehensive evaluation of patient education programs and activities 36 .
Timepoint: At one-month follow up.

Trial exit questionaaire
The trial exit questionnaire collected participant satisfaction, use of self-care, changes in understanding of menstruation, and information on visits to medical practitioners.