Introduction

Providing effective and holistic substance use disorder (SUD) prevention, treatment, and recovery services remains a major public health challenge in the United States and is of significant concern for maternal and child health populations. Perinatal SUD too often coincides with punitive measures and criminalization, rather than the evidence-based treatment proven to effectively support recovery, such as medication for opioid use disorder (MOUD) and peer support (Saia et al., 2016). Even when treatment is available, pregnant and parenting individuals experience lower rates of treatment retention due to time constraints, fear of child removal from custody, and stigma among healthcare providers (NIDA, 2020; Trainor, 2022).

Rural communities face unique challenges in addressing substance use disorders. Compared to urban areas, rural communities have higher overdose death rates (Centers for Disease Control & Prevention, 2017) and numerous barriers to providing comprehensive treatment. These barriers include a lack of transportation, limited employment opportunities, unaffordable housing, inadequate health insurance, and limited access to recovery support groups (Clark et al., 2021; Palombi et al., 2019). Additionally, the limited medical infrastructure in rural areas coupled with the stigmatization of substance use disorders makes MOUD difficult to access (Dombrowski et al., 2016). These challenges are particularly prevalent in the Midwest, where much of the region is rural, and overdose deaths rates continue to increase despite declines in parts of the Northeast (Ahmad et al., 2022). While the opioid epidemic has drawn widespread attention and national resources, methamphetamine use has largely been overlooked as a public health concern and is growing in prevalence in the Midwest (Courtney & Ray, 2014; Hedegaard et al., 2019). Unlike most other non-prescribed substances, methamphetamine use appears equally as prevalent in women as in men (Courtney & Ray, 2014).

The complexities of substance use among both perinatal populations and Midwestern populations demand an innovative approach to better understand interactions between perinatal people who use substances, public health systems, community-rooted services, and clinical providers. Applying a systems perspective to perinatal substance use allows for diverse stakeholders to collectively produce insights on the underlying dynamics driving outcomes, with the goal of informing strategic planning and policy efforts (Hovmand, 2014; Naumann et al., 2021; Sterman, 2000).

The National Maternal and Child Health (MCH) Workforce Development Center, a technical assistance center supported by the federal Maternal and Child Health Bureau (MCHB) since 2013, is uniquely positioned to accelerate strategic planning utilizing systems thinking tools (Coffey et al., 2022). The Workforce Development Center has over nine years of experience responding to emerging needs through tailored trainings and workshops for Title V programs and their partners in three core areas: systems integration, change management/adaptive leadership, and evidence-based decision-making.

In this paper, we highlight findings from a system-strengthening process facilitated by the National MCH Workforce Development Center with members and potential partners of the MidwestFootnote 1 Substance Use in Pregnancy Club (Midwest SUPper Club). The Midwest SUPper Club is a multi-disciplinary, grassroots regional coalition of perinatal and substance use providers, state public health agencies, and community-rooted organizations. The coalition formed in 2020 to address the need for Midwest-specific interventions, knowledge sharing, and collaboration (Zaman et al., 2023).

Methods

Qualitative group model building (GMB) and key informant interviews were used to identify opportunities for the Midwest SUPper Club to effectively leverage their collective expertise to advance outcomes for people with perinatal SUD. GMB is a community-based applied systems thinking methodology to elicit systems structures, stakeholder mental models,Footnote 2 and insights about how to meaningfully improve system behavior (Hovmand, 2014; Muttalib et al., 2021; Sterman, 2000). GMB differs from other stakeholder engagement methods in its use of facilitated, scripted activities to apply evidence-based components of systems thinking, as well as its strong emphasis on exploring potential positive and negative short- and longer-term impacts of policy and intervention options (Hovmand et al., 2012). GMB activity scripts are utilized to support better implementation of the GMB method and ensure fidelity to the process (Hovmand et al., 2012). In addition to the system-level insights produced through GMB, the process itself has the capacity to strengthen stakeholder collaboration by generating buy-in and providing a shared language for understanding the system. The MCH Workforce Development Center received Institutional Review Board approval for workforce development activities through the host university (IRB #14-0768).

Group Model Building

Midwest SUPper Club members and potential partners were invited to take part in a 2 day (10 h) virtual GMB workshop. Recruitment efforts were pragmatic, relying on SUPper Club leadershipFootnote 3 to extend personal invitations to existing partners and, when prompted by the workshop facilitators, to seek out additional attendees from state public health agencies, community-based organizations, and individuals with lived experience. A small stipend was offered for people with lived experience to attend. Recruitment intentionally centered around including perspectives from diverse sectors in the broad perinatal SUD treatment system in the Midwest. Twenty participants from four states (Iowa, Michigan Minnesota, and Wisconsin) participated in at least one day of the workshop, which reflects the Midwestern states in which the SUPper Club has had the highest level of engagement. Collectively, the organizations and individuals participating in the workshop brought rich insight into unique challenges of many sub-populations seeking out perinatal SUD care in the Midwest, including geographic, education-level, cultural, and racial/ethnic sub-populations. Roughly half of the participants were clinicians from various hospital/health-care systems (N = 10), with the other half representing public health agencies or organizations (N = 9) or lived experience of perinatal substance use disorder (N = 1). Due to scheduling conflicts, 19 participants attended Day 1 and 10 participants (9 returning and 1 new) attended Day 2.

Three facilitators trained in GMB planned and led the workshop following a series of scripted GMB activities described in Table 1 that were modified for the virtual context (Andersen & Richardson, 1997; Chin et al., 2022; Hovmand et al., 2012, 2015). The workshop was facilitated via Zoom, using both Google Slides and Vensim (a causal mapping and simulation software) to support activities. With permission, main session discussions were recorded for internal note-taking purposes and to capture meaningful quotes from participants. Following Day 1, in the three-day interval between workshops, facilitators reviewed the recording and all annotated Google Slides in order to develop final version(s) of the causal loop diagrams (CLDs). On Day 2, facilitators introduced the updated CLDs and initiated small group discussions to capture missing elements and insights. Finally, leverage points and corresponding priorities for action were elicited from the CLDs by participants.

Table 1 Description of group model building workshop activities
Full size table

Semi-Structured Interviews

Pairs of interviewers trained in GMB conducted semi-structured interviews with two subject matter experts from Iowa and Wisconsin whose schedules did not allow for them to attend the workshop. Both individuals were health care providers with a combined 14 years of experience directly serving perinatal people with SUD, and had been identified by SUPper Club leadership as having an important perspective to share. Each interview lasted approximately 60 min and was conducted virtually via Zoom. Interviewers first presented a summary of workshop deliverables, specifically focusing on the CLDs, and then asked for feedback to corroborate findings from the GMB workshop and continue refining the CLDs.

Workshop Evaluation

Following the workshop, participants were asked to complete a 22-question evaluation via Qualtrics. This evaluation was created by a member of the evaluation team at the MCH Workforce Development Center. The short survey, which was distributed to participants via an emailed survey link, included both multiple choice and open-ended questions on the workshop’s impact on system thinking skills at an individual and team level.

Results

Behavior-Over-Time Graphs

In two groups, participants developed a total of 20 behavior-over-time graphs (BOTG) during the workshop, capturing meaningful trends that the SUPper Club could focus change efforts on. The groups reached consensus through iterative discussion on two primary trends that capture the club’s role in strengthening the perinatal SUD system (presented in the Supplementary Material) : (1) Birthing peoples’ perception/experience of stigma from the treatment community, which was viewed as “high stigma” in the past and present, with a hope to decrease to “low or no stigma” in the future; and (2) The Midwest SUPper Club’s reach and influence, which has demonstrated deliberate growth since its inception in 2020, with a hope to avoid stagnation and continue to expand engagement in the future.

Causal Loop Diagrams

The three CLDs that emerged to capture separate but interrelated dynamics include: (1) the influence of stigma on maternal and infant health outcomes, (2) the role of clinic, organizational, and state policies, and (3) the impact of workforce education and evidence-based practices on care. See Tables 2, 3, 4 for detailed descriptions of the feedback loops identified in each model (including guidance on how to read and interpret CLDs) and the Supplementary Material for a list of endogenous and exogenous variables elicited.

Table 2 Feedback loops in the stigma and MCH outcomes model
Full size table
Table 3 Feedback loops in the policy model
Full size table
Table 4 Feedback loops in the workforce and evidence-based practices model
Full size table

The first CLD (Fig. 1, with larger diagrams available in the Supplementary Material) illustrates the many ways in which stigma impacts maternal and infant health outcomes. Participants discussed various levels of stigma and the differing effect of internalized stigma versus stigmatization from a provider or a system-at-large. The diagram represents these dynamics in three variables: (1) Self-stigma: an acceptance of a negative narrative about the self; (2) Social stigma: individual stereotypes, prejudice, or discrimination (consciously or subconsciously), towards an already stigmatized group; and, (3) Structural stigma: policies and culture within institutions that promote negative attitudes towards a group, either directly or indirectly (Livingston et al., 2012; Trainor, 2022).

Fig. 1
figure 1

Causal loop diagrams of the Midwest SUPper Club’s role in the perinatal SUD system. Bold, italicized text was used to indicate the variables that were identified in BOTGs as the priority trends the group should focus on (“SUPper Club reach and influence”, “birthing person’s perceptions/experience of stigma from the tx community”). Green was used to indicate variables that the SUPper Club believes they can directly impact through the work of their coalition. Orange boxes were used to identify variables that the SUPper Club strives to indirectly impact through the work of their coalition. Red dotted lines indicate where participants determined the SUPper club should prioritize future efforts and potential pathways for the SUPper club’s work to most significantly influence the system. CLD’s identify relationships between variables using labeled arrows which indicate that a change in the first variable triggers a change in the second variable over time, all else equal. The polarity of causal links is labeled “S” to indicate the variables change in the same direction (e.g., if the value of the first goes up, the value of the second also goes up); they are labeled “O” to indicate that the variables change in opposite directions (e.g., if the value of the first goes up, the value of the second goes down) (Color figure online)

Full size image

The second diagram (Fig. 1, with larger diagrams available in the Supplementary Material) resulted from rich discussions of the harmful impact of punitive policies and cross-system value and service alignment on perinatal SUD care. Timely connection to treatment and support emerged as a key outcome of interest. This requires both individuals’ basic needs (access to safe and supportive social, physical, and community environments) to be met as well as non-punitive evidence-based care (including screening/testing) to be provided. Connection to adequate treatment was found to be hindered by the fundamental system structure, which is not designed to engage or reach people who need perinatal SUD care. One health care provider shared an example of this: they reported hesitancy to screen patients for perinatal substance use due to inadequate referral options if use was identified. Additionally, a lack of awareness of treatment resources among perinatal people and their families, paired with fear of punitive consequences, leads to less timely connection to treatment and support.

The third diagram (Fig. 1, with larger diagrams available in the Supplementary Material) explores the impact of workforce education and evidence-based practices on the delivery of equitable, quality care. Workforce education in this context encompasses education on the impact of stigma, general knowledge on perinatal SUD, and information on available resources/referral sources and relevant laws and policies. Workforce education was emphasized for its potential to reduce stigma, increase systems alignment, and ultimately improve the quality of care delivered. The need for evidence-based practices for perinatal SUD was discussed broadly and in relation to caring for patients experiencing perinatal methamphetamine use. Unlike other types of substance use disorders such as opioid use disorder, methamphetamine poses significant challenges for treatment as it has few evidence-based treatment options (Courtney & Ray, 2014). It was also noted that the existence of evidence-based practices alone is not enough to influence the system—best practices must be made available through accessible training opportunities, supported at the systems- and organizational-level, and implemented alongside social and structural stigma reduction efforts.

After the workshop and interviews, the facilitation team developed a summary (Fig. 2) of system structure flaws and mental models (both problematic and desired) that emerged using the Iceberg Model of Systems Thinking (Maani & Cavana, 2007). The figure was validated by the SUPper club leadership and shared with all members. This summary is intended to support the SUPper Club in communicating and synthesizing key insights from the GMB process to stakeholders unfamiliar with CLDs.

Fig. 2
figure 2

Synthesized GMB insights mapped to the iceberg framework. The Iceberg Framework is a common model used in systems thinking to represent the visible part of the problem above the water line, and the underlying trends, structure, and mental models under the water line (Maani & Cavana, 2007). This figure summarizes findings from the GMB workshop and interviews

Full size image

Leverage Points for Action

Based on the CLDs, participants developed and prioritized leverage points for action, or places within the system the SUPper Club can have the greatest impact. Four priority areas emerged from these conversations: (1) alignment and promotion of shared mental models across stakeholders, (2) expansion of workforce education and training opportunities for the perinatal SUD workforce, (3) strengthening of infrastructure to support care navigation for patients and providers, and (4) collaborative identification of evidence-based and regionally focused practices. Across all leverage points, there is a strong emphasis on centering patients and communities directly affected by perinatal SUD in all decision-making.

Workshop Evaluation

Of the 20 participants who attended the workshop, 11 completed the post-workshop evaluation. 75% of respondents reported that participation in the workshop increased collaboration across the Midwest SUPper Club “to a high degree.” All respondents (N = 11) reported that participation in the workshop supported to a “high degree” or “very high degree” the development of a shared mental model of the most pertinent challenges facing perinatal people with SUD in the Midwest. One participant noted that the workshop “…helped [the Midwest SUPper Club] navigate what feels like an overwhelming amount of areas for us to work on, and instead focus on what people think is important for our organization.” Another participant reflected that the workshop “…really made me think about all the details of the work we do, the families and communities it affects, and the groundwork we need to accomplish before moving towards external funding support. This is definitely the motivation I needed to take the next step in our development.”

Discussion

The Influence of Stigma on Maternal and Infant Health Outcomes

The workshop and interviews highlighted ways in which stigmatization harms perinatal people with SUD and their infants and families, including through decreased access to services, care engagement, and effects on the overall quality of care provided (Patrick et al., 2017; Stone, 2015). One participant called attention to a structural flaw in the system that encourages providers and stakeholders to prioritize either maternal or infant needs, even though there is a broad recognition that maternal and infant health outcomes are deeply interrelated (Lander et al., 2013; Patrick et al., 2017). This points to the need for equitable, family-centered, non-judgmental perinatal SUD care that is well-coordinated and aligned across health and social services. Through the GMB process, the SUPper club determined that one of their key roles as a collaborative is to decrease stigma through workforce trainings, community education, and system-alignment efforts.

The Role of Clinic, Organizational, and State Policies

Punitive policies create and uphold barriers to engagement in perinatal and SUD care. The term “punitive policies” was used by participants in the workshop to refer to both state-level policies (i.e., legislation classifying perinatal substance use as child abuse or neglect), and clinic/organizational policies (i.e., a mandate to report perinatal substance use to the criminal justice and/or child welfare system) that punish and/or criminalize perinatal substance use. Such policies deter pregnant or parenting people from seeking care and do not improve maternal or infant outcomes (Angelotta et al., 2016; Faherty et al., 2019; Patrick et al., 2017). Criminalization of perinatal SUD operationalizes the belief that substance use can be reduced through punitive policies, which reinforces a harmful and inequitable system structure. These policies set families up for failure (often creating vicious cycles—reinforcing feedback loops) and encourage family separation, particularly among low-income families of color. Participants discussed many ways in which shifting to a supportive system that promotes family unification can improve outcomes, including by reducing adverse childhood experiences and by increasing the likelihood of engagement in SUD treatment (Harp & Oser, 2018; Lander et al., 2013; Lollar, 2017).

Punitive policies also make it difficult for community systems (i.e., child protective services and housing providers) and medical systems (i.e., clinical providers and recovery programs) to align to support people affected by SUD during pregnancy. An example of this from the GMB workshop came from a public health professional who shared that often unrealistic expectations and requirements from child protective service agencies (i.e., mandating sobriety within a specific number of weeks) leads to family separation, temporarily or permanently. Identifying and eliminating these types of punitive policies was discussed as a major opportunity to stimulate meaningful cross-system alignment and collaboration.

The Impact of Workforce Education and Evidence-Based Practices on Care

The GMB process illuminated the specific ways in which the SUPper Club seeks to strengthen the perinatal SUD system through cross-sector workforce education and the growth and dissemination of evidence-based practices. Participants discussed the overwhelming challenge of beginning to provide care for perinatal people with SUD, especially in organizations or systems that lack the necessary structure to adequately support implementation of perinatal SUD evidence-based treatment. The SUPper Club clarified their ideal role as both coordinators of regional workforce education opportunities and “boots on the ground,” meaning that the club aims to identify and call attention to gaps in perinatal SUD evidence-based practices and to highlight opportunities for and generate further research. In addition to formal evidence-based practices, informal provider-to-provider consultation emerged as an important factor in delivering equitable, evidence-based care.

GMB as a Method to Accelerate Collaborative Planning Efforts

The GMB process, particularly CLD, is a powerful method to leverage diverse perspectives, equalize power dynamics and decision-making authority, visualize system structure, and collectively identify more influential leverage points for action. As the Midwest SUPper Club is in the early stages of development, GMB tools demonstrated a particular utility in building collaboration and reaching consensus across a broad range of stakeholder perspectives. For example, participants came in advocating for attention to issues nearest to their areas of expertise (e.g., “we too often neglect infants”), but left agreeing that stigma was a key fundamental issue to tackle first, and that conceptualizing their priorities as part of an inter-connected system strengthened all of their efforts. Further, these methods pushed participants to consider which perspectives were missing from the group. One participant reflected afterwards that “… since meeting with the MCH Workforce [Development Center], we were able to invite a member of the [local child protective services] team to attend our meetings, which broadens the perspective on this topic.” The SUPper Club plans to use the insights generated from this process to further grow and engage their membership and collaborate on best practice development, clinical outcomes research, and workforce learning opportunities (i.e., book clubs, annual conferences, and quarterly meetings).

This process has a few limitations. GMB workshops can be particularly challenging virtually and scheduling challenges led to non-homogenous participation across the 2 day workshop. Additionally, the virtual format required us to conduct our evaluation via a survey link rather than a paper survey; a limitation of our workshop was that only 55% of participants completed the evaluation survey. However, we feel we demonstrated scripted activities can still result in meaningful engagement in a virtual environment with a flexible participation structure. It should also be noted that although outreach was conducted to invite many individuals with lived experience to participate, ultimately only one person attended. This outreach was through personal contact and invitation of community-rooted organizations with whom individuals with lived experience had a trusted relationship.

GMB, specifically CLD, is reliant on skilled facilitators and takes practice to build competency. The GMB process is inherently iterative and should be enhanced with new perspectives and updated over time as the system changes. We support the on-going efforts in the system dynamics field to develop community-based GMB facilitators to inform and lead these efforts, and we believe that repeated exposure to systems thinking methodologies and utilization of the structured, scripted activities available can support the development of new facilitators (Hovmand, 2014; Hovmand et al., 2012, 2015). For this reason, the MCH Workforce Development Center prioritizes building systems thinking capacity and exposing MCH leaders, like those engaged in the SUPper Club, to the GMB process.

Future Directions

This GMB process helped to illuminate the complexity of the perinatal SUD system in the largely rural Midwest and adds to the body of literature exploring substance use in rural America. Although the findings from this GMB process are specific to the Midwest, we believe the themes extracted are applicable to other regions. This demonstrates an opportunity to further explore and validate system dynamics in other perinatal SUD contexts using GMB to illicit perspectives and reach consensus on action steps to strengthen the system. We also see opportunities for future research and practice directed towards better understanding the system factors producing growing inequities in overdose death rates across sub-populations. Black and Indigenous people are disproportionately affected by fatal overdose, and a deeper understanding of the complexity of the issue and possible interventions is warranted (Kariisa et al., 2022). Lastly, while GMB evaluations, including those described in this paper, often focus on evaluation of workshop processes and outcomes (i.e., changes in mental models), future research should evaluate the long-term impact and sustainability of GMB-generated solutions on strengthening MCH and behavioral health systems (Felmingham et al., 2023).

As demonstrated in the present paper, the GMB process offers a pathway that could be employed broadly across the MCH system to bring together new partners and support a shared understanding of the problem and specific actions to improve outcomes of interest over time (Cilenti et al., 2019). We see great potential for the expanded use of this method among state and local MCH, clinical, and community-rooted partners to accelerate system-strengthening efforts through intentional collaboration and shared decision-making.