https://orcid.org/0000-0003-3479-7591
Figures
Abstract
Perinatal mental illness is an important public health issue, with one in five birthing persons experiencing clinically significant symptoms of anxiety and/or depression during pregnancy or the postpartum period. The purpose of this study was to develop a consensus-based model of integrated perinatal mental health care to enhance service delivery and improve parent and family outcomes. We conducted a three-round Delphi study using online surveys to reach consensus (≥75% agreement) on key domains and indicators of integrated perinatal mental health care. We invited modifications to indicators and domains during each round and shared a summary of results with participants following rounds one and two. Descriptive statistics were generated for quantitative data and a thematic analysis of qualitative data was undertaken. Study participants included professional experts in perinatal mental health (e.g., clinicians, researchers) (n = 36) and people with lived experience of perinatal mental illness within the past 5 years from across Canada (e.g., patients, family members) (n = 11). Consensus was reached and all nine domains of the proposed model for integrated perinatal mental health care were retained. Qualitative results informed the modification of indicators and development of an additional domain and indicators capturing the need for antiracist, culturally safe care. The development of an integrated model of perinatal mental health benefitted from diverse expertise to guide the focus of included domains and indicators. Engaging in a consensus-building process helps to create the conditions for change within health services.
Citation: Ou C, Daly Z, Carter M, Hall WA, Zusman EZ, Russolillo A, et al. (2024) Developing consensus to enhance perinatal mental health through a model of integrated care: Delphi study. PLoS ONE 19(5): e0303012. https://doi.org/10.1371/journal.pone.0303012
Editor: Maria José Nogueira, School of Nursing Sao Joao de Deus, Evora University, PORTUGAL
Received: December 27, 2023; Accepted: April 17, 2024; Published: May 9, 2024
Copyright: © 2024 Ou et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Data cannot be shared publicly because of its potentially identifying and sensitive nature. Data are available from Wellstream: The Canadian Centre for Innovation in Child and Youth Mental Health and Substance Use (contact via wellstream.bienamont@ubc.ca) for researchers who meet the criteria for access to confidential data.
Funding: This study was funded by a Michael Smith Health Research British Columbia Convening & Collaborating Award grant (2020). The funding agency, Michael Smith Health Research BC (C2-2020-1426), had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Perinatal mental illness (PMI), which is mental illness experienced between the time of conception and the first year following childbirth, is an important public health issue. One in five birthing persons experience clinically significant symptoms of depression and/or anxiety during pregnancy or after giving birth [1–3] and an estimated one in 12 people report suicidal ideation [4]. Access to comprehensive and effective perinatal mental health (PMH) care is increasingly recognized as critical to improving long-term health and psychosocial outcomes for parents and their children [5–7].
Without access to holistic and coordinated PMH care, parents and children may experience profound health consequences. PMI is linked to adverse obstetric outcomes, such as preterm birth, low birth weight, small-for-gestational-age, stillbirth, and the need for operative deliveries [8,9]. PMI is also associated with adverse child outcomes, including developmental delays and socioemotional and behavioral challenges, such as attention deficit hyperactivity disorder [10–12]. Left untreated, PMI can become chronic, with widespread economic and social consequences [13,14]. Indeed, untreated or under-treated PMI is estimated to contribute substantial lifetime economic costs related to health and social care needs, and productivity and health-related quality of life losses [15,16]. In the US, untreated parental PMIs from 2017 alone were estimated to cost the system $14 billion from conception to 5 years after birth [17].
Internationally, screening and treatment for PMI is widely recommended, yet it remains unevenly executed [18]. Inconsistency in implementation results in a significant healthcare burden, including psychiatric admissions, outpatient treatment, and primary care visits [19]. In the Canadian context, access to clinically appropriate PMH care is hindered by the absence of routine screening for PMI and inadequately trained health care providers [20]. Even when symptoms are identified, systemic barriers to care prevent many individuals from receiving evidence-based treatment [21].
Despite high rates of PMI, the healthcare system is not equipped to adequately address care needs. A recent survey of perinatal care providers in Canada (n = 435) found that 96% of respondents believed that current PMH care services were insufficient to meet the needs of parents [20]. Moreover, the majority of care providers (87%) reported that many parents encounter significant barriers to accessing perinatal services, including limited availability, out-of-pocket costs, and a failure to address cultural safety [20]. As such, PMH services remain uncoordinated, with access often determined by ‘postal code lottery’, with those from urban, affluent, and well-resourced communities faring better than those from historically marginalized communities [20,22]. Because available services are often narrow in focus and ignore the broader social determinants of health—such as income and housing—pathways to screening, treatment, and follow-up tend to be unclear and difficult to navigate [20,23]. In the existing system, white, higher-income women with mild-to-moderate depression or anxiety often have better access to care and resources [22]. Parents who belong to historically underrepresented or marginalized groups, including rural residents, or those who are racialized, 2SLGBTQIA+, and/or living with complex mental illnesses, are particularly vulnerable to unmet PMH care needs [24,25].
Researchers and health professionals have argued that uncoordinated PMH services can be remedied by implementing an integrated model of PMH care. Integrated PMH care is the provision of coordinated, multidisciplinary care services—guided by evidence—to comprehensively address health and social care needs [5]. An integrated PMH care model can reduce system gaps and prevent the ‘screening paradox,’ wherein PMI is identified but left unmanaged because of a lack of referral pathways or resources [26]. An integrated PMH care model often uses a stepped care approach. That is, the least resource intensive care is offered first, ‘stepping up’ to progressively specialized, intensive PMH services as clinically indicated. Such a model is intended to make interdisciplinary and multimodal services available (i.e., coordinated medical, psychological, and pharmacological interventions) [6]. Primary care providers, many of whom do not have formal training in PMH care, can provide referrals, collaborate with PMH specialists, and access treatment records to support care trajectories [27,28]. Integrated PMH care is regarded as a health systems innovation that enhances perinatal mental health care equity by decreasing barriers to access and supporting the provision of coordinated care for all who need it based on best available evidence [28].
While integrated models of PMH care are considered promising for improving responses to population PMH care needs, there remains limited agreement on core model elements [19,29]. To address these knowledge and system gaps, this study aimed to generate a consensus-based model of integrated PMH care, with the potential to be applied in the Canadian health care context and beyond. Following, we detail our study using Spranger et al.’s reporting guidelines for Delphi techniques in health sciences [30].
Study context
The present study was conducted in British Columbia (BC), Canada; a province that operates in the context of a publicly-funded health care system. This system was developed to provide citizens with medically necessary hospital and physician services without out-of-pocket costs. However, the system, as organized, has been referred to as a two-tiered system when it comes to mental health care (i.e., many mental health services are not publicly-funded).
Each province/territory is responsible for structuring and maintaining a system of health care for its residents. Although BC is recognized as a Canadian leader in advancing PMH care [23], services remain siloed and there are barriers to care that include inadequate screening, uneven resource distribution, and long wait times [31]. There is an urgent need for research to inform new models of care to enhance integration of existing services.
Materials and methods
Study design
This study used a three-round Delphi design to inform a consensus-based model of integrated PMH care. To ensure the consideration of diverse perspectives, panel members included health care providers, researchers, and people with lived experience of PMI. Delphi methods use multi-stage panel surveys, which can include qualitative and quantitative data, to facilitate consensus building among experts [31]. Following each stage of surveying, feedback is provided to panel participants (e.g., aggregated survey results in the form of a short report). Panel participants are then provided with the opportunity to reconsider their responses in relation to this feedback and to indicate their evolving perspectives through subsequent survey rounds [31,32].
Delphi items can be generated through empirical evidence in combination with expert perspectives [33]. To identify core, evidence-informed domains (i.e., dimensions of integrated perinatal mental health care) and indicators (i.e., attributes of each dimension) [33], our study began with a scoping review of empirical literature (manuscript in-preparation). To populate a Delphi consensus-building survey, our diverse and interdisciplinary research team, which includes people with lived experience of PMI, used evidence from this scoping review to identify domains and indicators relevant to an integrated model of PMH care.
Participant identification and recruitment
Panel participants were recruited from across Canada and included perinatal clinicians and service providers (e.g., nurses, midwives, physicians, doulas, psychologists, pharmacists, social workers), researchers, and PMI patients and family members. Professional experts (i.e., clinicians, researchers, service providers) were eligible to participate if they had a minimum of two years of experience working in the context of PMH care. Patient and family experts were eligible to participate if they had experiences with PMI within the past five years. We identified a diverse group of subject experts resulting in an initial list of potential participants (n = 94). Prospective panel members were provided with the study purpose and invited to complete a series of confidential Delphi consensus-building surveys consisting of queries with pre-determined response options (i.e., quantitative) and open-ended questions (i.e., qualitative). Surveys were created in Qualtrics and a survey link was included in the invitation email, with prospective participants encouraged to share the invitation within their networks to facilitate recruitment. Ethics approval for this study was obtained from the University of British Columbia Behavioural Research Ethics Board (H21-00651). Participants completed an online consent form prior to initiating the survey and were provided with an honoraria of $25 CAD (gift card) in Rounds 1 and 2, and $50 CAD (gift card) in Round 3 to improve retention.
Data collection
Delphi Round 1.
The survey link was emailed to prospective participants in October 2021. Participants were asked to evaluate the indicators in each identified domain for relevance (i.e., whether the indicator contributes to an integrated model of PMH care), importance (i.e., the value of the indicator to an integrated model of PMH care), and relative rank (i.e., the comparative importance of the indicator relative to other indicators). Comments and suggested modifications for the proposed domains and indicators were also invited. Participants were asked to assess the relevance of each indicator with the response options: “Keep,” “Modify,” or “Remove.” Participants were asked to assess importance with the response options: “Not at all important,” “Somewhat important,” “Very important,” and “Extremely important.” Participants were asked to rank indicators in the order of greatest to least importance. Based on common practice in Delphi surveying, a threshold of ≥ 75% agreement across participants was considered to indicate consensus on the relevance and importance of assessed indicators [34]. We also sought open-ended feedback on the domains and indicators, as well as input on content that participants perceived as missing from the model. The first round of surveying included 44 indicators across nine domains, and our primary goal was to assess consensus for retaining the proposed indicators. Participants were given a 2-week window to submit the survey and received reminders to encourage completion.
Delphi Round 2.
Following Round 1 surveying, we summarized the results, including qualitative feedback, and provided a summary report to all participants. In December 2021, a second survey was emailed to participants who were asked to re-assess each domain and indicator while taking into consideration other participants’ input. Our primary goal for this round was to collect additional qualitative data to inform potential modifications. All participants from Round 1 were invited and sent weekly reminders to complete the survey within the two-week timeframe.
Delphi Round 3.
Following Round 2 surveying, we summarized the results and provided a summary report to all participants. In response to Round 2 findings, which identified the need for another domain and additional indicators, a third round of surveying was initiated in April 2022. Based on conversations with study partners who work in health services administration, we invited 16 additional participants to Round 3 to bring further diversity, specifically the perspectives of Indigenous and 2SLGBTQIA+ people. Participants in Round 3 were again asked to provide information on the relevance, importance, and relative rank of the modified and retained items as well as the added domain and related indicators. Comments about further potential modifications and general perceptions were invited. All participants from Round 2 as well as the new Round 3 participants were sent weekly reminders to complete the survey within the two-week timeframe.
Data analysis
All quantitative data were uploaded to SPSS version 28.0 to facilitate analysis. Descriptive statistics were used to characterize the sample and to assess consensus on the relevance and importance of domains and indicators across survey rounds. To examine relative rank, we drew on Round 3 data and used the Borda Count method to produce a summary score. The Borda Count is a scoring method where indicators are awarded points according to their rankings [35]. Where participants skipped the ranking process within a domain, they were excluded from analysis. If participants partially completed the ranking (e.g., only assigned ranks to their top three indicators), the unranked indicators were allotted a score of one [35]. Qualitative data were analyzed according to survey round, domain of interest, and indicator examined. Using an inductive approach, data were clustered into subthemes and higher order themes, establishing links between participant responses.
Results
Participant overview
Of the 94 prospective participants invited, 31 participated in Round 1. Of these, a total of 20 returned to participate in Round 2. In Round 3, 13 participants returned from Round 2 and two participants returned from Round 1. Additionally, 16 new participants were recruited for a total of 31 Round 3 participants. Across all three rounds, the majority of participants identified as women/female, lived in a metropolitan area, and had attended/completed post-secondary education. The majority of participants were clinicians or direct care providers, with approximately half of participants residing in British Columbia. The proportion of participants who were PMH care service users (i.e., patients or family members) ranged from 16% in Round 1 to 29% in Round 3 (see Table 1).
Quantitative results
Ratings of relevance and importance are provided in Table 2. Consensus was reached for the relevance of all indicators within each domain across all three rounds (i.e., 44 indicators across 9 domains in Rounds 1 and 2, and 57 indicators across 10 domains in Round 3). The number of indicators reaching consensus (≥ 75% agreement) as being “extremely important” or “very important” to an integrated model of PMH care varied across rounds: 43 of 44 indicators reached concensus in Round 1, 37 of 44 indicators reached consensus in Round 2, and 53 of 57 indicators reached concensus in Round 3. The four indicators that did not achieve consensus as being “extremely important” or “very important” in the third and final round were: indicators 4 and 6 in the “Person-centred care” domain; indicator 5 in the “Transition and discharge planning” domain; and indicator 3 in the “Care model planning” domain (see Table 3). Furthermore, in Round 3, 4 of 57 indicators achieved consensus on being “extremely important”. These were indicators 3 and 5 in the “Health promotion and illness prevention” domain, and indicators 3 and 5 in the “Biopsychosocial approach to treatment” domain (see Table 3). The relative rankings of indicators using Borda count methods are provided in Table 2. None of the four indicators that reached consensus on being extremely important were the highest ranked in their respective domains. Indicators that failed to reach consensus on importance were also the lowest ranked in their respective domains (see Table 2).
Qualitative results
Although the quantitative data indicated a high level of agreement on the relevance, and to a lesser extent the importance of indicators, qualitative data from the open-ended responses provided a more nuanced picture. In total, four themes were developed from the qualitative data: representation of care recipients; care recipient advocacy; language for cultural safety, cultural humility, and anti-racism; and incorporation of systemic approaches.
Participants who responded with modifications for indicators and suggested additional indicators, emphasized the importance of attention to representation of care recipients (Theme 1). There were two subthemes. The first was rejection of the term ‘patient’. Participants who objected to using patient as a descriptor offered alternatives, such as persons, people, or clients. For example, “people rather than patients” was a sentiment provided frequently in Round 1. The second subtheme was family inclusion. Participants consistently commented about including family, as defined by the care recipient, in the indicators for domains, including person-centred care, care delivery, health promotion and prevention, biopsychosocial approach to treatment, and transition and discharge planning. This included comments like, “Health promotion activities with the whole family unit” (Round 1) and “I think preconception planning could also include relationship support and co-parenting plans, since this seems to be associated with people staying well over time.” (Round 2).
The second theme suggested indicator modifications and additional indicators related to care recipient advocacy. In all three rounds, participants emphasized the importance of starting from the perspective of the care recipient and care recipients advocating for themselves. They argued that adequate information about options for care was fundamental to care recipients’ efforts to collaborate with care providers by taking the lead and expressing preferences (choices). For example, “As a patient, I didn’t know what to expect. Just being asked "what do you need?" isn’t always enough when we aren’t necessarily aware of what options exist. I think more guidance around what is doable within the patient’s goals (Round 1)”. There were two subthemes: providing options to support collaboration and supporting care recipients’ choices. Providing options to support collaboration, represented a number of comments about the importance of information. For example, “if the client doesn’t know options, unable to lead” (Round 1) and “takes the lead after an informed choice discussion” (Round 2). In the supporting choices subtheme, participants wrote how indicators could be altered. For example, “all patients are offered routine screening–need to ensure patient autonomy and choice to be screened” (Round 3) and “providing an option for self-referral or self-advocacy for clients…” (Round 1).
The third theme was language for cultural safety, cultural humility, and anti-racism. Many participants responded to the new domain, identified in Round 3. Participants had consistently written about the need for attention to cultural safety in Rounds 1 and 2. For example, “the care team collaborates with patients and their families to develop individualized care plans that are trauma-informed and culturally sensitive” (Round 1) and “in a culturally safe way” (Round 2). Participants paid close attention to Round 3 indicator language. There were three subthemes: White supremacy, objections to BIPOC, and inadequate team power. The first subtheme was supported by lengthy comments. Participants indicated that White supremacy and ‘current’ racism in Canada required acknowledgement. For example, “the care team practices cultural humility, a process of life-long self-reflection to examine personal and systemic biases, assumptions, and power imbalances that serve to maintain the status quo of White supremacy and marginalization of Black, Indigenous, and People of Colour (BIPOC).” Some of the participants objected to the use of BIPOC. They regarded the term as limiting and suggested that it excluded marginalized groups, such as immigrant populations. For example, “Indigenous first, avoid acronyms as they can change and are limiting.” The third subtheme was inadequate team power. Some participants expressed concerns that care teams did not have the wherewithal to make the changes spelled out in the indicators. For example, with regards to culturally safe services, one participant wrote, “I don’t think you can put this on the care team. It may not be up to them. I think that you need a different level of indicators for the people in charge.” With regards to calls to action from the Truth and Reconcilliation Commission of Canada, which sought to reflect the stories of those who were impacted directly or indirectly by the Indian Residential Schools system in Canada, another participant wrote, “The care team might not have that power”.
The fourth and final theme was: incorporation of systemic approaches. Some participants expressed concern about the model operating at an ‘individual’ level. For example, participants described indicators that could capture the importance of support and social determinants of health in health promotion: “Place weight on in-home and community supports, socio-economic status, access to other resources as a factor in triage beyond individual mental health indicators” (Round 1).
An important subtheme was the universality of screening for PMIs. Participants objected to limiting screening to particular settings and groups. For example, “I think the word ‘universal’ must be included in screening, it can be done in multiple settings, [Canada Prenatal Nutrition Program] and other maternity programs, on pregnancy, postpartum and lactation apps…” (Round 1) and “Screening may happen in settings with a trained health care professional who is aware of next steps, have had education and are aware of resources in health care and community” (Round 2).
In the second subtheme, participants wanted treatment and transition plans to attend to social determinants of health, barriers to treatment, and stakeholder engagement. They were concerned about silos rather than integration of community and acute care settings. For example, “Ongoing assessment and monitoring includes: treatment effectiveness and potential treatment modifications, changes in social determinants of health for the client and approaches to meeting fundamental needs…, psychosocial and relational risk factors…, and collaborative problem-solving around child care support/psychosocial needs” (Round 3) and “The treatment plan is trauma-informed, culturally sensitive, and responsive to patients’ values, preferences and needs…” (Round 1).
In the third subtheme, the participants called for national components to support a perinatal mental health model. They wanted standardization. For example, “National infrastructure developed as a central resource for up-to-date educational offerings and maintaining/curating educational resources (a "one-stop shop" for those working in PMH care to stay current)” (Round 1).
In general, comments about the model indicated a preference for more continuity of care from various settings; separating out each treatment component and changing the weighting of indicators to actual treatment choices; simplifying the language; and reducing the clinical focus of the model. Participants’ acknowledged leadership in thought and time devoted to the survey; attention to trauma-informed care; intersectionality with BIPOC and Indigenous issues; and attention to culturally competent care.
The themes and general comments provided context for the changes made in indicators and domains (e.g., new label for person-centered care instead of patient-centred care, new domain for cultural safety, cultural humility, and anti-racism) for Round 3 of the Delphi Survey.
Summary of changes to the model during the delphi process and resulting final version.
In Rounds 1 and 2, the model presented to participants was identical. In response to participants’ comments in Rounds 1 and 2, modifications and additions were made to the model that was assessed in Round 3, including the addition of a new domain, new indicators, and modified indicators. The final version of the model, including all domains and indicators is presented in Table 3.
Discussion
There is growing recognition of the need to improve accessibility and coordination of evidence-based PMH care to enhance parent and family outcomes. Despite awareness of current gaps in services, the essential components of an integrated model of PMH care remain unclear. In this study, our interdisciplinary team proposed a model of integrated PMH care developed through a review of evidence designed to address this gap in knowledge. Using Delphi methods, we then sought input from a diverse group of stakeholders, including clinicians, researchers and people with lived experience of PMI in order to gain consensus on key model components. Implementing the resulting model may enhance the quality, accessibility, and effectiveness of PMH care services, thereby improving outcomes for parents and families.
While there are profound siloes that continue to serve as a barrier to good care and outcomes for patients navigating PMI, we achieved evidence-aligned consensus among experts and people with lived experience of PMIs for the key elements of an integrated model of PMH care. Research on integrated PMH care services has underscored the importance of coordinated and scaffolded services that feature inter-professional collaboration, shared vision and goals, continuity of care, and adequate funding and resources to support coordination [36,37]. Indeed, evidence from systematic reviews synthesizing research on non-perinatal integrated models of care indicate that they improve care quality and access, while also lowering operational costs [38] and improving patient satisfaction [38,39]. In the perinatal setting, preliminary evaluation of a state-wide integrated model of PMH care in the US demonstrated improvements in mental health outcomes and patient satisfaction [40]. Qualitative findings from our study extend these findings and indicate the need for centralized processes to support care coordination, referrals, and education. Indeed, adopting a national strategy to provide a coordinated approach to PMH care holds the potential to harmonize screening practices and care standards across provinces and territories, while also enhancing integration and access to care through centralized funding and resource allocation [20,22,23].
Methodologically, the Delphi process emphasized the need for careful consideration of data used to determine the acceptability and sufficiency of a proposed model of care. Indeed, while Delphi methods provide an approach to building consensus on issues of interest in health contexts, relying solely on the collection of quantitative data risks prematurely–and inaccurately–declaring consensus as having been achieved. In the context of this study, the quantitative data provided an incomplete picture of the adequacy and comprehensiveness of the proposed integrated model of PMH care. For example, consensus on the relevance of indicators was achieved in all three rounds of surveying, and the majority of indicators reached consensus in relation to importance. If the team relied exclusively on quantitative data, this study may have been considered complete after Round 1 surveying. Qualitative input suggested that several indicators required modification and that an additional domain was required. This feedback was carefully considered by the study team and informed the creation of the domain of “Cultural safety, cultural humility, and anti-racism”, as well as the alteration and addition of several indicators. Indeed, a mixed methods approach provided critical insights and new understandings that would not have been identified if the study relied solely on quantitative data. Qualitative data invites diverse perspectives and expertise to be shared to bring richer insights and support a collaborative process where panel participants can more meaningfully inform the model or framework generated [41,42]. The utility of collecting both qualitative and quantitative data in Delphi studies has been noted by others who have observed that a mixed methods approach is essential for identifying framework deficiencies and necessary modifications [41–43].
Beyond the importance of using mixed methods, this study benefited from attention to the characteristics and composition of the participant panel. For example, minority perspectives risk being overshadowed when relying on numerical consensus methods alone. As such, efforts were made to enhance participant heterogeneity through the purposive sampling of participants who identified as belonging to an equity-deserving group, such as recent immigrants and people who identify as 2SLGBTQIA+. While the attrition of participants from Round 1 and the introduction of new participants in Round 3 may have reduced consensus on the importance of certain indicators, inviting new participants to Round 3 facilitated a diversity of perspectives in assessing the new domain of “Cultural safety, cultural humility, and anti-racism” and related indicators. As noted by others, there is an inherent tension that needs to be managed between facilitating sample homogeneity to reach consensus and maintaining enough heterogeneity to detect model gaps [44].
The consensus-based model of integrated PMH care generated through this study provides key insights that build on the existing research in this field. Importantly, our panel participants highlighted the need for greater emphasis on culturally safe care in PMH programming. This omission was identified by several participants in Rounds 1 and 2 through qualitative feedback. The addition of the cultural safety, cultural humility, and anti-racism domain addresses a critical gap in much of the existing research on PMH care. This domain acknowledges that childbearing people from equity-deserving groups (e.g., racialized and Indigenous peoples) routinely encounter racism and discrimination in health care [45,46]. Racism and discrimination are embedded in organizational structures of health care in many Western contexts, resulting in inequities in maternal and infant morbidity and mortality among certain population groups [41,47,48]. Moreover, socioeconomic disparities often compound racial inequities [48,49]. For example, research has demonstrated that racialized women are less likely to be screened for depression [50] and to receive treatment [51]. The addition of this 10th domain is aligned with recent efforts to recognize and redress issues of equity and justice in health care access and treatment for populations who experience structural vulnerabliity [52,53].
Strengths & limitations
This study provides new insights and a consensus-based model of integrated PMH care informed by a geographically, ethnoculturally, and experientially diverse participant panel; however, there are limitations that warrant discussion. To present the various domains and indicators of an integrated model of PMH care, the Delphi survey was long and participation required a substantial time commitment. The time commitment likely contributed to attrition following the Round 1 survey. Efforts were made to expand the panel and more appropriately acknowledge contributions through a larger honoraria offered in Round 3. Data collection issues arising from the question structure on the Qualtrics platform precluded the use of ranking data from Rounds 1 and 2. That problem holds implications for insights into the relative importance of the indicators. We resolved the issue in Round 3, by providing an indication of relative importance of indicators in this final survey. Finally, while the qualitative data provided important findings that led to the creation of a new domain as well as related indicators, the collection of these data through open-ended survey questions limits the opportunity for more nuanced exploration or validation of interpretations. Future research would benefit from the incorporation of qualitative focus groups to further refine and finalize model components.
Clinical implications
This study developed a consensus-based model of integrated PMH care that holds the potential to guide health services redesign to improve PMH care treatment, access and patient experience [54]. Future research should include pilot testing to assess the feasibility and acceptability of this model in diverse care contexts [55]. Data on process and outcome measures and qualitative feedback from service users and clinicians can inform model refinements to improve usability and effectiveness [54]. Model implementation feasibility metrics–such as cost-effectiveness and stakeholders’ willingness to adopt the model–have utility to determine the model’s practicality, scalability, and potential for wider application [54,55].
Conclusion
Perinatal mental illness is a key public health issue requiring innovation in health services planning and delivery to improve outcomes for parents and their children across the life course. Using Delphi methods, this study contributed to of the development of a consensus-based, evidence-aligned and “expert”-informed model of integrated PMH care to guide health systems and practice change. Grounded in scientific evidence the model was strengthened by the incorporation of expert and lived experience knowledge to include 10 priority domains: person-centred care; cultural safety; care delivery; health promotion; screening; biopsychosocial treatment; transition support; education; strategic care model planning; and evaluation. Future research is needed to support implementation and testing of this model in diverse settings.
Acknowledgments
The authors are thankful for the contributions of the people and families with lived experience of perinatal mental illness whose perspectives were instrumental to refining our integrated model of care. We also thank Dr. Brittany Bingham (Vancouver Coastal Health), Dr. Valerie Rychel (St. Paul’s Hospital), Dr. Nichole Fairbrother (University of British Columbia) for their assistance with increasing the diversity of our participant panel as well as Liza McGuinness (University of British Columbia) and Rajveen Sharma (Vancouver Coastal Health) for supporting the development of the Qualtrics survey.
References
- 1. Wang G, Deng Y, Jiang Y, Lin Q, Dong S, Song Y, et al. Trajectories of sleep quality from late pregnancy to 36 months postpartum and association with maternal mood disturbances: a longitudinal and prospective cohort study. Sleep. 2018;41. pmid:30535183
- 2. Dennis C-L, Falah-Hassani K, Shiri R. Prevalence of antenatal and postnatal anxiety: systematic review and meta-analysis. The British Journal of Psychiatry. 2017;210: 315–323. pmid:28302701
- 3. Yin X, Sun N, Jiang N, Xu X, Gan Y, Zhang J, et al. Prevalence and associated factors of antenatal depression: Systematic reviews and meta-analyses. Clinical psychology review. 2021;83: 101932. pmid:33176244
- 4. Xiao M, Hu Y, Huang S, Wang G, Zhao J, Lei J. Prevalence of suicidal ideation in pregnancy and the postpartum: A systematic review and meta-analysis. Journal of affective disorders. 2022;296: 322–336. pmid:34600967
- 5. Lomonaco-Haycraft KC, Hyer J, Tibbits B, Grote J, Stainback-Tracy K, Ulrickson C, et al. Integrated perinatal mental health care: a national model of perinatal primary care in vulnerable populations. Primary Health Care Research & Development. 2019;20: e77. pmid:29911521
- 6. Lim I, Newman-Morris V, Hill R, Hoehn E, Kowalenko N, Matacz R, et al. You can’t have one without the other: The case for integrated perinatal and infant mental health services. Aust N Z J Psychiatry. 2022;56: 586–588. pmid:35257590
- 7. Kruper A, Wichman C. Integrated Perinatal Mental Health Care. Psychiatric Annals. 2017;47: 368–373.
- 8. Bussières E-L, Tarabulsy GM, Pearson J, Tessier R, Forest J-C, Giguère Y. Maternal prenatal stress and infant birth weight and gestational age: A meta-analysis of prospective studies. Developmental Review. 2015;36: 179–199.
- 9. Jahan N, Went TR, Sultan W, Sapkota A, Khurshid H, Qureshi IA, et al. Untreated Depression During Pregnancy and Its Effect on Pregnancy Outcomes: A Systematic Review. Cureus. 13: e17251. pmid:34540477
- 10. Manzari N, Matvienko-Sikar K, Baldoni F, O’Keeffe GW, Khashan AS. Prenatal maternal stress and risk of neurodevelopmental disorders in the offspring: a systematic review and meta-analysis. Soc Psychiatry Psychiatr Epidemiol. 2019;54: 1299–1309. pmid:31324962
- 11. Madigan S, Oatley H, Racine N, Fearon RMP, Schumacher L, Akbari E, et al. A Meta-Analysis of Maternal Prenatal Depression and Anxiety on Child Socioemotional Development. Journal of the American Academy of Child & Adolescent Psychiatry. 2018;57: 645–657.e8. pmid:30196868
- 12. Glover V. Maternal depression, anxiety and stress during pregnancy and child outcome; what needs to be done. Best Practice & Research Clinical Obstetrics & Gynaecology. 2014;28: 25–35. pmid:24090740
- 13. Rokicki S, McGovern M, Von Jaglinsky A, Reichman NE. Depression in the postpartum year and life course economic trajectories. American Journal of Preventive Medicine. 2022;62: 165–173. pmid:34696940
- 14. Meltzer-Brody S, Stuebe A. The long-term psychiatric and medical prognosis of perinatal mental illness. Best Practice & Research Clinical Obstetrics & Gynaecology. 2014;28: 49–60. pmid:24063973
- 15. Bauer A, Knapp M, Parsonage M. Lifetime costs of perinatal anxiety and depression. Journal of affective disorders. 2016;192: 83–90. pmid:26707352
- 16. Bauer A, Pawlby S, Plant DT, King D, Pariante CM, Knapp M. Perinatal depression and child development: exploring the economic consequences from a South London cohort. Psychological medicine. 2015;45: 51–61. pmid:25066467
- 17. Luca DL, Margiotta C, Staatz C, Garlow E, Christensen A, Zivin K. Financial Toll of Untreated Perinatal Mood and Anxiety Disorders Among 2017 Births in the United States. Am J Public Health. 2020;110: 888–896. pmid:32298167
- 18. Griffen A, McIntyre L, Belsito JZ, Burkhard J, Davis W, Kimmel M, et al. Perinatal Mental Health Care In The United States: An Overview Of Policies And Programs: Study examines perinatal mental health care policies and programs in the United States. Health Affairs. 2021;40: 1543–1550.
- 19. Howard LM, Khalifeh H. Perinatal mental health: a review of progress and challenges. World Psychiatry. 2020;19: 313–327. pmid:32931106
- 20. Hicks LM, Ou C, Charlebois J, Tarasoff L, Pawluski J, Roos LE, et al. Assessment of Canadian perinatal mental health services from the provider perspective: Where can we improve? Frontiers in psychiatry. 2022;13: 929496. pmid:36213903
- 21. Vanderkruik R, Freeman MP, Nonacs R, Jellinek M, Gaw ML, Clifford CA, et al. To screen or not to screen: Are we asking the right question? In response to considering de-implementation of universal perinatal depression screening. General Hospital Psychiatry. 2023;83: 81–85. pmid:37141774
- 22. DeRoche C, Hooykaas A, Ou C, Charlebois J, King K. Examining the gaps in perinatal mental health care: A qualitative study of the perceptions of perinatal service providers in Canada. Frontiers in Global Women’s Health. 2023;4: 1027409. pmid:37009091
- 23. Hippman CL, Adham M, Zlobin C, Wong G. An Environmental Scan of Perinatal Mental Health Infrastructure Across Canada. J Obstet Gynaecol Can. 2022;44: 745–746. pmid:35589519
- 24. Declercq E, Feinberg E, Belanoff C. Racial inequities in the course of treating perinatal mental health challenges: Results from listening to mothers in California. Birth. 2022;49: 132–140. pmid:34459012
- 25. Singla DR, Meltzer-Brody S, Savel K, Silver RK. Scaling up patient-centered psychological treatments for perinatal depression in the wake of a global pandemic. Frontiers in psychiatry. 2022;12: 826019. pmid:35197873
- 26. Chrzan-Dętkoś M, Walczak-Kozłowska T. Postpartum depression crisis since the second lockdown and ‘screening paradox’: many women identified, very few treated. BMC Public Health. 2022;22: 2387. pmid:36539811
- 27. Branquinho M, Shakeel N, Horsch A, Fonseca A. Frontline health professionals’ perinatal depression literacy: A systematic review. Midwifery. 2022;111: 103365. pmid:35640360
- 28. Hansen MED, Tobón AL, Haider UK, Simas TAM, Newsome M, Finelli J, et al. The role of perinatal psychiatry access programs in advancing mental health equity. General Hospital Psychiatry. Elsevier; 2023.
- 29. Manolova G, Waqas A, Chowdhary N, Salisbury TT, Dua T. Integrating perinatal mental healthcare into maternal and perinatal services in low and middle income countries. bmj. 2023;381.
- 30. Spranger J, Homberg A, Sonnberger M, Niederberger M. Reporting guidelines for Delphi techniques in health sciences: A methodological review. Zeitschrift für Evidenz, Fortbildung und Qualität im Gesundheitswesen. 2022;172: 1–11. pmid:35718726
- 31.
Keeney S, McKenna HA, Hasson F. The Delphi technique in nursing and health research. John Wiley & Sons; 2011.
- 32. Van der Steen JT, Radbruch L, Hertogh CM, de Boer ME, Hughes JC, Larkin P, et al. White paper defining optimal palliative care in older people with dementia: a Delphi study and recommendations from the European Association for Palliative Care. Palliative medicine. 2014;28: 197–209. pmid:23828874
- 33. Freitas Â, Santana P, Oliveira MD, Almendra R, Bana e Costa JC, Bana e Costa CA. Indicators for evaluating European population health: a Delphi selection process. BMC Public Health. 2018;18: 557. pmid:29703176
- 34. Diamond IR, Grant RC, Feldman BM, Pencharz PB, Ling SC, Moore AM, et al. Defining consensus: A systematic review recommends methodologic criteria for reporting of Delphi studies. Journal of Clinical Epidemiology. 2014;67: 401–409. pmid:24581294
- 35. Ehringfeld I, Steinhoff P, Störk S, Whittal A, Herber OR. The Voting Scheme Determines the Winner. Considering Possible Options for Designing a Delphi Study. International Journal of Qualitative Methods. 2023;22: 16094069221148871.
- 36. D’haenens F, Van Rompaey B, Swinnen E, Dilles T, Beeckman K. The effects of continuity of care on the health of mother and child in the postnatal period: a systematic review. European journal of public health. 2020;30: 749–760. pmid:31121019
- 37. Coates D, Coppleson D, Travaglia J. Factors supporting the implementation of integrated care between physical and mental health services: an integrative review. Journal of Interprofessional Care. 2022;36: 245–258. pmid:33438489
- 38. Rocks S, Berntson D, Gil-Salmerón A, Kadu M, Ehrenberg N, Stein V, et al. Cost and effects of integrated care: a systematic literature review and meta-analysis. Eur J Health Econ. 2020;21: 1211–1221. pmid:32632820
- 39. Baxter S, Johnson M, Chambers D, Sutton A, Goyder E, Booth A. The effects of integrated care: a systematic review of UK and international evidence. BMC Health Services Research. 2018;18: 350. pmid:29747651
- 40. Muzik M, Menke RA, Issa M, Fisk C, Charles J, Jester JM. Evaluation of the Michigan Clinical Consultation and Care Program: An Evidence-Based Approach to Perinatal Mental Healthcare. Journal of Clinical Medicine. 2023;12: 4836. pmid:37510951
- 41. Clark E, Vedam S, Mclean A, Stoll K, Lo W, Hall WA. Using the Delphi method to validate indicators of respectful maternity care for high resource countries. Journal of Nursing Measurement. 2022. pmid:35705228
- 42. Rowell LL, Polush EY, Riel M, Bruewer A. Action researchers’ perspectives about the distinguishing characteristics of action research: a Delphi and learning circles mixed-methods study. Educational action research. 2015;23: 243–270.
- 43. Hardin KY, Black C, Caldbick K, Kelly M, Malhotra A, Tidd C, et al. Current practices among speech-language pathologists for mild traumatic brain injury: A mixed-methods modified Delphi approach. American Journal of Speech-Language Pathology. 2021;30: 1625–1655. pmid:34038183
- 44. Belton I, MacDonald A, Wright G, Hamlin I. Improving the practical application of the Delphi method in group-based judgment: A six-step prescription for a well-founded and defensible process. Technological Forecasting and Social Change. 2019;147: 72–82.
- 45. OjiNjideka Hemphill N, Crooks N, Zhang W, Fitter F, Erbe K, Rutherford JN, et al. Obstetric experiences of young black mothers: An intersectional perspective. Social Science & Medicine. 2023;317: 115604. pmid:36549014
- 46. Bacciaglia M, Neufeld HT, Neiterman E, Krishnan A, Johnston S, Wright K. Indigenous maternal health and health services within Canada: a scoping review. BMC Pregnancy Childbirth. 2023;23: 327. pmid:37158865
- 47. Glazer KB, Howell EA. A way forward in the maternal mortality crisis: addressing maternal health disparities and mental health. Arch Womens Ment Health. 2021;24: 823–830. pmid:34427773
- 48. Huggins B, Jones C, Adeyinka O, Ofomata A, Drake C, Kondas C. Racial Disparities in Perinatal Mental Health. Psychiatric Annals. 2020;50: 489–493.
- 49. Craemer KA, Garland CE, Sayah L, Duffecy J, Geller SE, Maki PM. Perinatal mental health in low-income urban and rural patients: The importance of screening for comorbidities. General Hospital Psychiatry. 2023;83: 130–139. pmid:37187032
- 50. Sidebottom A, Vacquier M, LaRusso E, Erickson D, Hardeman R. Perinatal depression screening practices in a large health system: identifying current state and assessing opportunities to provide more equitable care. Archives of women’s mental health. 2021;24: 133–144. pmid:32372299
- 51. Salameh TN, Hall LA, Crawford TN, Staten RR, Hall MT. Racial/ethnic differences in mental health treatment among a national sample of pregnant women with mental health and/or substance use disorders in the United States. Journal of psychosomatic research. 2019;121: 74–80. pmid:30928211
- 52. Webb R, Ayers S, Shakespeare J. Improving accessing to perinatal mental health care. Journal of Reproductive and Infant Psychology. 2022;40: 435–438. pmid:36285387
- 53. Ayers S, Crawley R, Alderdice F, Eagle A. Personalised approaches to intervention for perinatal mental health difficulties. Journal of Reproductive and Infant Psychology. 2021;39: 339–341. pmid:34406105
- 54. Moore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process evaluation of complex interventions: Medical Research Council guidance. BMJ. 2015;350: h1258. pmid:25791983
- 55. Pearson N, Naylor P-J, Ashe MC, Fernandez M, Yoong SL, Wolfenden L. Guidance for conducting feasibility and pilot studies for implementation trials. Pilot Feasibility Stud. 2020;6: 167. pmid:33292770