OptimalMe is a type 3 hybrid effectiveness-implementation study [35] with an active intervention phase (2 years) and a passive observation phase (up to 5 years). Type 3 hybrid implementation designs are those in which implementation outcomes are primary, and individual or population outcomes are secondary [35]. The primary outcome of the project includes overall intervention penetration and reach and the feasibility, acceptability, adoption, and fidelity of the delivery of OptimalMe. Secondary outcomes include evaluation of individual health outcomes associated with implementation delivery mode, including knowledge, risk perception, health literacy, self-management, and health behaviors. The study design is a parallel, two-arm, randomized trial at the level of the individual utilizing a pragmatic philosophy, working within real-world conditions to assess overall effectiveness. All individuals will receive the same evidence-based lifestyle intervention, and implementation delivery methods will be compared.

The Australian health care system is government supported via Medicare, which provides universal free care to Australian citizens and residents (and others who are eligible) and is supported by a subsidized private health system. Private health insurance, paid by the individual, allows patients to choose hospitals and health care providers from outside of the public system, with a 12-month waiting period before some, or all, of the cost of hospital treatment is reimbursed. Overall, approximately 54% of Australian adults have a form of private health insurance [36], and 27% of births in Australia occur in private hospitals [37]. Women who give birth in private hospitals attend ambulatory private obstetric care and have limited contact with hospitals prior to delivery.

Women who upgrade private insurance to include, or join with, pregnancy and birth coverage, comprise a unique population, prospectively signaling future intention for a pregnancy. In a sample of 294 women who had recently upgraded or obtained insurance for pregnancy and birth coverage, 41% intended to conceive within the next 12 months [33].

The implementation partner in this research program is Medibank Private, which is one of Australia’s largest insurers (funding approximately 20,000 births annually). Feasibility scoping shows approximately 7800 women join with, or upgrade to, pregnancy and birth coverage with this insurer nationally each year.

This implementation research is underpinned by the Consolidated Framework for Implementation Research [38], which provides a pragmatic framework, informed by translation into practice theories, that is designed to guide complex implementation projects and generate knowledge across settings and studies [38]. The framework consists of 5 domains [38]: Domain 1 consists of the unadapted intervention to be implemented and assumes the intervention is composed of core or fundamental components, essential to efficacy, surrounded by peripheral components that are adaptable to the local context, without altering the integrity of the intervention. The adaptable components are informed by domain 2 (the outer setting, ie, policy, guidelines, population needs), domain 3 (the inner setting, ie, the organization’s structure, culture, readiness to change), and domain 4 (the individuals within the outer or inner setting involved in the intervention as influencers of implementation. The implementation process (domain 5) works across all domains to achieve implementation through an iterative change process of executing and evaluating implementation activities [38].

The fundamental core components of our intervention include theoretical underpinning; simple diet; physical activity and self-management messages; low-intensity delivery format; individual health coaching sessions focused on goal setting, problem solving, and self-management delivered by a qualified health professional; and ongoing intervention support via text messaging (Domain 1). Core components were informed by our extensive intervention evidence base [16,20,21,23] and were consistently applied to setting, population, intervention tools, resources, and delivery method and format. The integration of peripheral intervention components was undertaken to incorporate best practice clinical guidelines and systematic review lifestyle intervention evidence (Domain 2) and using an intervention co-design process with the implementation partner, Medibank Private (Domain 3) experts in obstetric and lifestyle delivery and reproductive women in this life stage (Domain 4). This included the incorporation of health education resources within the intervention and the development of a consumer-tested web-based digital platform for remote delivery and comanaged participant engagement. The efficacy of health coaching delivery methods (phone and videoconference) will be compared. A governance process has been established to enable responsive and pragmatic adaptations to peripheral components in partnership with the implementation partner (Medibank Private), yet designed and managed by the clinician academic research group (Monash University). The primary outcomes form Domain 5, which includes overall intervention feasibility, reach, acceptability, and adoption as well as fidelity of the delivery of OptimalMe as planned.

The study will be conducted in accordance with Consolidated Standards of Reporting Trials [39] and Template for Intervention Description and Replication frameworks [40].

The target intervention population will include Medibank Private members who have joined or upgraded with pregnancy and birth coverage within the 3 months prior to recruitment (to align with likely planned conception based on insurance uptake and wait times). Eligibility criteria focus on inclusiveness and includes those who are not pregnant, who wish to conceive within 12 months of recruitment, who are aged 18 to 44 years, with any BMI, who read and speak English, and who have access to an internet-capable device will be included.

Given the implementation effectiveness study design, sample size has not been powered on a clinical outcome because the primary outcome is to determine implementation learnings to inform feasibility. Available funding enables intervention delivery to approximately 300 women, which is approximately 10% of the eligible population with intention to conceive, based on our formative research [33].

Participants will be randomized to receive health coaching either by telephone or via videoconference. An external senior statistician will provide computer-generated randomization codes to the research coordinator only, who will sequentially allocate all participants but will have no role in intervention delivery. Researchers involved in intervention delivery and data collection will be blinded to the allocation sequence; however, due to the nature of the intervention, they will not be blinded to participant allocation. Researchers responsible for data analysis and reporting will be blinded to both allocation sequence and participant allocation. Randomization performed external to the implementation setting is designed to reduce bias. Due to the nature of the intervention, participants will not be blinded to group allocation.

A co-designed process, using an opt-in design, was developed with the implementation partner to facilitate Australia-wide recruitment. We will use direct email communication (approximately 500 members every month, to be varied based on response rates and historical trends in email engagement observed during specified periods, including seasonal holiday periods) to recruit women (randomly selected to receive a targeted invitation by system generated mailing lists) who meet initial eligibility criteria (insurance coverage and age) with a link to the landing page of the web-based intervention platform. The page contains introductory information about the healthy lifestyle intervention, including a video. Individuals who wish to take part will be required to confirm remaining eligibility criteria, including pregnancy status and intention to conceive within 12 months, provide informed consent electronically, and register via a digital interface. The researcher coordinator will then contact the Medibank-managed integrated voice recognition system to confirm the potential participant’s membership (using first and last name, membership ID, date of birth, and postal code) and pregnancy and birth coverage status to confirm eligibility. An email with an activation link to an account for the intervention will be sent to participants, at which time, they are randomized to 1 of the 2 coaching delivery methods. Recruitment will continue until target numbers (n=300) are reached. This pragmatic approach enables management of participant flow into the intervention and will not disqualify those who may be unaware of pregnancy status at point of recruitment or who may re-evaluate their intention to become pregnant after recruitment (Multimedia Appendix 1).

This intervention is based on our previous healthy lifestyle program—HeLP-her [16,20-23,27]). HeLP-her is a low intensity, behavior change program, grounded in social cognitive theory [41]. HeLP-her is nonprescriptive and provides health coach–delivered simple messages on healthy lifestyle behaviors aligned with national dietary and physical activity guidelines [30,42,43]. These are reinforced by behavior change strategies including identifying individual health priorities and facilitators and barriers for change. Realistic achievable goals are prioritized and developed by participants, and a behavioral action plan that outlines goals and timeframes is established. Individual barriers, strategies for change, and social supports are identified and discussed, and self-monitoring is practiced and encouraged [27].

Co-design of intervention delivery with the implementation partner prioritized remote delivery to ensure equitable accessibility to the intervention across Australia by using a dedicated web-based digital platform, supported by health coaching (delivered via phone or videoconference), with ongoing text message support (Figure 1).

The secure web-based platform can be accessed via desktop and mobile apps. The platform contains preconception, pregnancy, and postpartum modules. Participants are provided access to the preconception module at the outset from the dashboard (Figure 2). The pregnancy module is accessible to participants when they update their personal profile on the dashboard (pregnancy status and estimated due date). In addition to the 3 modules, the dashboard contains an interactive BMI (and gestational weight gain, if pregnancy is reported) calculator, a checklist, and activities to review behavioral action plans.

All modules have a similar format—each has a health information (education) section and a healthy lifestyle behavior change (self-management) section. The health information section contains health, medical, and screening information and the healthy lifestyle section contains diet and physical activity recommendations, as well as an interactive behavior change section. Health, medical, and screening information are presented as a suite of fact sheets specific to each reproductive phase (Multimedia Appendix 2). Information provided is based on the Royal Australian College of General Practitioners Red Book [29], Australian Government Clinical Practice Guidelines for Pregnancy [30], and our formative research [33,34]. A series of health screening questions at module commencement based on these guidelines will inform the presentation of fact sheets according to relevance. For example, preconception participants will be asked when their last cervical screen was completed. If a participant indicates a cervical screen outside of a guideline-specified timeframe, the relevant cervical screening fact sheet will be presented under Essential information at the top of the screen. Conversely, if a participant indicates a cervical screen within a guideline-specified timeframe, the fact sheet will be presented under Other recommended reading. This design feature ensures that participants are directed to the information that is most relevant to their health needs (based on their responses to screening questions) while minimizing the burden of information, which has previously been identified as a barrier to receiving health information [31].

Fact sheets have a similar structure—each fact sheet has 3 to 4 key messages, followed by detailed topic information and links to other websites for additional evidence-based information. Each fact sheet is supported by an interactive component that enables the user to populate a checklist item summarized on the platform dashboard (Figure 2). For example, opening a fact sheet about cervical screening will populate the Check my cervical screening status with my GP item. Health literacy is tested at the top of each fact sheet with a true or false question, with corresponding information that explains the correct answer.

Healthy lifestyle resources include fact sheets related to weight gain prevention and infographics that are based on Australian adult pregnancy [30] and breastfeeding [42] dietary guidelines and physical activity guidelines [43]. Additional resources include information on how to read food labels, how to estimate food portion sizes, healthy snack and food substitution ideas, and calories consumed using various food choices versus those burned from walking. Behavior change is supported by an interactive goal-setting section that guides the user to develop a personalized goal through action planning, which includes an activity to self-select modifiable health behaviors for improvement (eg, packaged or convenience food consumption, alcohol intake, physical activity, fruit and vegetable intake, and sleep). Areas of relevance are selected and prioritized by each participant (ie, areas are ranked in order of importance). Participants are guided through goal-setting using free text to specify what they would like to achieve identify motivating factors and social support pathways, barriers to behavior change and specific strategies to overcome barriers that are time dependent and identification. An action item to review goals will be automatically added to the digital platform, which is encouraged 2 weeks after goal planning and commitment (Figure 2).

The platform was consumer-tested using a quantitative survey for functionality (ie, ease of navigation across the platform), acceptability (ie, usefulness of the information, presentation and aesthetics, ease of understanding content) and relevance (ie, appropriateness of information, potential for the platform to assist in optimizing health behaviors, peer recommendation). The survey contained a series of statements requiring response on a 5-point Likert scale that ranged from strongly disagree to strongly agree, with an opportunity to provide free text. Responses were transformed to a binary representation (0, disagree and neutral; 1, agree). Overall, 36 women were recruited from the community with advertisements across all modules using both computer (19/36, 53%) and mobile phone or tablet (17/36, 47%) devices. Women were aged between 25 and 38 years old, and the majority were university educated (33/36, 92%). Most women agreed that they could navigate to different areas of the platform and return to the dashboard with ease (23/30, 77%); they found the platform to be aesthetically appealing (25/36, 69%), and the amount of information to be appropriate (28/36, 78%) and easy to understand in its presentation (29/30, 97%). The majority believed that the information would assist them in improving their health-related behaviors (25/30, 83%) and considered the platform relevant to recommend to women of the same life stage (24/36, 67%).

Health coaches with a tertiary qualification in health sciences (ie, dietetics, nutrition, exercise physiology, or allied health) will deliver the program and aim for continuity wherever possible to maintain rapport. The purpose of the individual health coaching sessions is to build rapport with participants, reiterate program objectives, enhance engagement, practice goal setting and self-management skills, support participants with lifestyle modules, and provide personalized feedback on behavior change. Any module components that have not been accessed or completed by participants at the point of health coaching will be flagged for completion by the participant during or after the session.

All participants will be offered 2 personalized preconception health coaching sessions (approximately 20 minutes in duration, delivered either be phone or videoconference according to randomization) 2 to 4 weeks after program entry and 6 weeks later.

During pregnancy, 2 additional 20-minute health coaching sessions will be scheduled (8-10 weeks gestation or 1-2 weeks after starting the pregnancy module commencement and 19-20 weeks gestation).

SMS text messages will be sent every 3 weeks as a reminder to practice healthy behaviors.

The economic evaluation will be designed to identify the costs associated with implementation, and the net costs to health care funders. Costs of the OptimalMe implementation package will be identified from the trial data, including the costs of platform maintenance, staff time (in providing the coaching sessions), and SMS text messages. Fixed and variable costs will be identified, allowing cost per woman to be estimated at different scales of the intervention. The net costs to health care funders will be identified by quantifying the costs associated with birth type, birth complications, neonatal intensive care unit admissions, hospital stay, and ancillary utilization. Costs to Medicare Benefits Schedule will be identified based upon item numbers [51] associated with birth type and complications. Costs to private health insurers associated with hospital stay will be identified directly from the study. Costs to public hospital funders from neonatal intensive care unit admissions or any public hospital transfers will be identified from the National Hospital Cost Data Collection produced by the Independent Hospital Pricing Authority [52]. The total cost per woman will be calculated, and generalized linear models will be utilized to identify differences in costs between delivery methods. We will use these models to estimate the net cost impacts to different funders at different levels of the population reached.