Miscarriage Australia: the use of a human centered design approach to design and develop a website for those affected by miscarriage

Background

Past research has shown that Australians affected by miscarriage want a website specific to both miscarriage and their local region that is accessible, comprehensive, evidence-based and informed by experts. The aim of this study was to design, develop and evaluate the Miscarriage Australia website using human centered design.

Methods

A four stage human centered design approach was used to develop the Miscarriage Australia website which aimed to: (1) Understand the issue and why users need a website; (2) Define users' specific needs; (3) Design solutions to meet those needs; and (4) Evaluate the design by testing with end users. Across the four stages, various types of data and data analysis were developed and utilized including interviews, desktop research, development of personas and tone of voice, followed by usability testing. Process and content were guided by designers, developers and an expert advisory committee of key stakeholders.

Results

Analysis and synthesis of user research across Stages 1 and Stage 2 found 11 key themes pertaining to user's miscarriage experiences and support needs. Using the themes, common experiences, goals, motivations and behaviors of users were identified and similar user types grouped and used to inform the development of two personas. Using the personas and user research findings, design elements (Stage 3) including the "tone of voice guidelines" were developed recommending the Miscarriage Australia website be calm, empathetic, hopeful and authoritative. The tone of voice guidelines guided branding and over 100 pages of content was informed by the research team and reviewed by a 13-member Expert Clinical Advisory Committee over two rounds to ensure it was evidence based and reflected best practice. Using a contextual inquiry approach, usability testing was undertaken with 8 end users to test a low fidelity mockup and high-fidelity prototype of the website. Overall, end users reported the website was highly acceptable in terms of the design, content, layout, language and terminology, describing it in line with the intended tone of voice. Users reported the website was easy to use and navigate and provided useful and appropriate content and resources. Minor areas for improvement included slight changes to specific images, improved links for navigating sections, and a title change to one section heading.

Conclusion

The Miscarriage Australia website was successfully implemented and commended by users as meeting their needs. As a result of using human centered design, the Miscarriage Australia website provides an ideal template or blueprint on how to develop a successful and useful digital resource for users, particularly around sensitive women's health issues.

Heart Rate Variability Indices as Possible Biomarkers for the Severity of Post-traumatic Stress Disorder Following Pregnancy Loss

Background: Psychological distress, such as posttraumatic stress disorder (PTSD), is commonly evaluated using subjective questionnaires, a method prone to self-report bias. The study's working hypothesis was that levels of autonomic dysfunction determined by heart rate variability (HRV) measures are associated with the severity of PTSD in women following pregnancy loss. Methods: This was an observational prospective cohort study with 53 patients enrolled. The DSM-5 (Diagnostic and Statistical Manual of Mental Disorders) PTSD scale (PCL-5) was used to assess the severity of PTSD in women after pregnancy loss. The cardiac autonomic function was assessed using HRV measurements during a deep breathing test using an HRV scanner system with wireless ECG enabling real-time data analysis and visualization. HRV measures were: standard deviation (SD) of normal R-R wave intervals [SDNN, ms], square root of the mean of the sum of the squares of differences between adjacent normal R wave intervals [RMSSD, ms], and the number of all R-R intervals in which the change in consecutive normal sinus intervals exceeds 50 milliseconds divided by the total number of R-R intervals measured [pNN50 = (NN50/n-1)^*100%] [pNN50%]. Results: The PCL-5 scores had a statistically significant association with HRV indices (SDNN; RMSSD, and pNN50%). Patients with PTSD had similar mean heart rate values as compared to patients without PTSD (PCL-5), but significantly higher SDNN [median[IQR, interquartile range]: 90.1 (69.1-112.1) vs. 52.5 (36.8-65.6)], RMSSD [59.4 (37.5-74.9) vs. 31.9 (19.3 - 44.0)], and PNN50% values [25.7 (16.4-37.7) vs. 10.6 (1.5-21.9)]. The SDNN of the deep breathing test HRV was effective at distinguishing between patients with PTSD and those without, with an AUC = 0.83 +/- 0.06 (95 % CI 0.94, p = 0.0001) of the ROC model. Conclusions: In this study, HRV indices as biomarkers of cardiac dysautonomia were found to be significantly related to the severity of PTSD symptoms in women after pregnancy loss.