Abstract
Introduction
Designing novel solutions to address care fragmentation in the perinatal ecosystem requires rigorous evaluation of current workflows and engagement of the wide net of stakeholders supporting patient care. Human‐centered design (HCD) is an established methodology for identifying challenges faced by patients and providers in clinical workflows, with the goal of producing actionable insights to inform interventions. This study used an HCD approach to (1) understand transitions of perinatal care within a safety‐net system and (2) identify barriers that informed design opportunities to improve care coordination for socially complex pregnant patients.
Methods
We conducted semi‐structured interviews with patients, clinical providers (physicians, midwives, nurses, social workers), and community‐based organization (CBO) staff working in a safety‐net healthcare system. Interviews were analyzed using an HCD approach to inductive thematic analysis.
Results
Thirty‐four interviews were conducted with 42 participants, including 13 patients, 22 clinical providers, and seven community‐based or public health stakeholders. We developed six insight statements, with themes centered around fragmented communication, limited integration between clinical and community systems, uncertainty surrounding information‐sharing policies, and inconsistent implementation of substance‐use‐related protocols. Participants also highlighted disruptions to continuity of care and variability in discharge planning that contributed to confusion and eroded trust in the care team.
Conclusions
Using an HCD approach, we identified key challenges in care coordination for socially complex perinatal patients. Interventions that seek to strengthen informational continuity between clinical and community‐based providers offer promise for improved perinatal care delivery in safety‐net systems.
Keywords: care coordination, human‐centered design, informational continuity, prenatal care, safety‐net hospital, transitions of care
1. INTRODUCTION
The care needs of pregnant patients are multifaceted and often extend beyond the direct clinical care provided by obstetricians. Coordinating care for pregnant patients can be a significant challenge for perinatal providers, especially those serving patients with complex social needs [1]. At safety‐net hospitals, which primarily serve low‐income people, patients frequently experience financial and logistical barriers to accessing perinatal care. Limited clinic hours, transportation issues, job insecurity, provider shortages, and underinsurance make accessing care challenging [2, 3]. Patients with complex social needs, including those experiencing homelessness [4, 5], substance use disorders [6, 7], incarceration [8], intimate partner violence [9, 10], and mental illness, face additional barriers to accessing pregnancy services. As a result, prenatal care utilization is lower among these high‐risk populations [11].
Pregnant patients seeking care at safety‐net hospitals often receive services from a variety of clinical sites, public welfare programs, and community‐based organizations (CBOs). Siloed services and the vast number of distinct providers and departments involved in one patient's care can result in the loss of continuity [12, 13, 14]. “Informational continuity,” defined as the “use of information on past events and personal circumstances to make current care appropriate for each individual,” is central to highly functioning care systems [13, 14]. When informational continuity is lost across transitions of care, patient experiences are negatively affected. Losing informational continuity jeopardizes patients’ trust in their providers and the healthcare system, contributing to higher rates of care avoidance [13]. Care fragmentation also burdens healthcare providers, who are engaged in duplicate documentation and parallel care processes. Most importantly, pregnant people in fragmented care systems experience poorer clinical outcomes [1, 15].
Various institutions have piloted solutions to improve informational continuity in obstetric settings. Efforts have been made to improve transitions of care through communication procedures, such as standardized handoffs for hospital admissions or transfers [16, 17]. Multidisciplinary time‐outs have also been tested as a method to decrease information lost at key clinical decision points [17]. However, these solutions are aimed at improving team communication in situations in which providers are already in dialogue, rather than facilitating new methods of information sharing across patients’ pregnancy journeys. In safety‐net hospitals specifically, institutions have often relied on case managers to coordinate care for socially complex patients [18, 19, 20, 21]. While effective, this high‐touch solution is challenging to scale as it requires significant resources that are often unavailable or difficult to access in under‐resourced practice settings. Solutions that are sustainable given the financial and personnel constraints of most public hospital systems are required to improve care coordination. Ultimately, improving the quality of care around times of transition will result in improved patient care experiences and outcomes.
Designing novel solutions to address care fragmentation in the perinatal ecosystem requires rigorous evaluation of current workflows and engagement of the wide net of stakeholders supporting patient care. Human‐centered design (HCD) is a well‐established methodology that offers a framework for identifying problems in existing workflows, with the goal of producing actionable insights to inform possible interventions. Employing ethnographic research methods, HCD facilitates a deep understanding of users’ unmet needs. The HCD process also includes rigorous methods for iteratively prototyping solutions to address these identified needs, with the goal of developing and refining sustainable solutions [22]. Importantly, the HCD methodology is rooted in empathy toward users to ensure that design solutions address real problems and build trust among important stakeholders [22, 23, 24].
The objectives of this study were to use an HCD approach to (1) understand the process of transitioning care within the perinatal care system and (2) identify scalable opportunities to improve care coordination for patients receiving perinatal services at urban safety‐net hospitals. In this study, we define transitions of care as the movement of patients and information between providers, settings, or phases of the perinatal continuum (e.g., prenatal, intrapartum, and postpartum care). We define care coordination as the processes and procedures that facilitate these transitions, including delivery of healthcare services, exchange of information, and delineation of provider roles and responsibilities. By focusing on transitions of care (Objective 1), we sought to identify where breakdowns occur during periods of shifting care responsibility. By examining care coordination (Objective 2), we aimed to identify system‐level opportunities to strengthen communication and alignment across clinical and community‐based care settings.
2. METHODS AND MATERIALS
2.1. Setting
This prospective observational study was conducted at San Francisco General Hospital, an urban safety‐net hospital in San Francisco. The study was designed and implemented as a collaboration between the Solid Start Initiative, a multidisciplinary collaboration aimed at improving the health and well‐being of young families, and The Better Lab, a mixed‐methods research center specializing in HCD research techniques.
2.2. Human‐centered design
The HCD methodology breaks down the process of solutions design into three distinct phases. The purpose of the first phase, “Inspiration,” is to develop a deep understanding of the needs and challenges faced by stakeholders and to identify a set of insights and opportunity areas that address their unmet needs [24]. Throughout the inspiration phase, users and designers engage in a “synthesis” process—a collaborative approach to analysis in HCD studies—where findings are condensed into meaningful and actionable “insights” (termed “insight statements”) to inform future solution‐generation. Insight statements summarize key user needs, motivations, and tensions revealed through the data analysis/synthesis process. These insights are then translated into design opportunities—actionable opportunity statements that capture potential areas for intervention. The purpose of the second phase, “Ideation,” is to use the insights statement and design opportunities developed in the Inspiration phase to generate a breadth of ideas that can be refined into potential solutions that can be prototyped [24]. Finally, the purpose of the third phase, “Implementation,” is to develop prototypes of the concepts created in the “Ideation” phase and to test the outputs, with the goal of understanding the approval, fit, and feasibility of implementation (Figure 1) [24, 25].
FIGURE 1.
Phases of the human‐centered design process—inspiration, ideation, and implementation—with corresponding research and design activities.
In this study, we focus on the “Inspiration” phase of the HCD process, with the goal of generating insights and opportunities for interventions to improve care coordination within perinatal ecosystem. To identify challenges within current workflows and areas of user need, we conducted semi‐structured interviews with patients and both clinical and community‐based stakeholders.
2.3. Data collection
Purposeful sampling was used to identify patients in perinatal care settings with varying levels of social need, stages of perinatal journey (prenatal to 3 years postpartum), racial/ethnic backgrounds, and levels of care engagement to capture a diversity of care experiences. This included 13 patient interviews, comprising eight interviews with postpartum patients and five interviews with members of our Community Advisory Board (a group of current and former perinatal patients that advise on patient‐centered initiatives in the safety‐net system). Patients who were not members of the Community Advisory Board were approached during inpatient admissions to Labor and Delivery and offered participation in the study.
Purposeful sampling was also used to identify providers and key stakeholders working in the safety‐net healthcare system, CBOs, and the Department of Public Health to represent the breadth of care services and professional experiences. Participants were recruited through professional networks of clinical providers on the study team (C.N.S., M.T., N.D., H.B., M.S., and L.O.), clinical workspaces, and well‐established perinatal programs across the city. Following interviews, participants were also asked to suggest additional key stakeholder to invite to participate in the study. In total, 22 perinatal providers and staff participated in 17 interviews with individual perinatal providers (one to three participants per interview), and seven community‐based and public health stakeholders participated in four interviews, at which point thematic saturation was reached. Individual versus group interview formats were selected based on the nature of the stakeholder group and practical considerations of participation. Individual interviews were used for patients and with most clinical providers to support confidentiality and allow participants to describe personal experiences in greater depth. Group interviews were used for CBO staff and staff working closely together (e.g., in the same role/capacity) to encourage discussion of shared workflows. The composition of individual and group interviews is summarized in Table 1.
TABLE 1.
Interviewed stakeholders by setting and role.
| Setting | Interviews (n) | Participants (n) | Participant roles |
|---|---|---|---|
| Perinatal care stakeholders | |||
| Labor and delivery unit | 8 | 11 | Nursing staff/leadership (n = 3), midwifery (n = 1), social work (n = 1), attending pediatrician (n = 1), attending obstetrician (n = 1), medical students engaged in quality improvement initiatives on the unit (n = 4, group interview) |
| Outpatient obstetrics clinic—clinical staff | 2 | 4 | Attending obstetrician (n = 1), family medicine attendings (n = 3, group interview) |
| Outpatient obstetrics clinic—support staff | 3 | 3 | Practice manager (n = 1), health worker (n = 1), social worker (n = 1) |
| Obstetric‐psychiatry clinic | 4 | 4 | Attending psychiatrist (n = 1), psychiatry resident physician (n = 1), behavioral health specialist (n = 1), program evaluation specialist (n = 1) |
| Community‐based and public health stakeholders | |||
| Community‐based organizations (CBOs) | 3 | 6 | Program managers (n = 2), program staff (n = 4) from three CBOs (group interviews) |
| Public health nurse | 1 | 1 | Public health nurse (n = 1) |
| Patients | |||
| Patients | 13 | 13 | Postpartum patients (n = 8), patient members of the Community Advisory Board (n = 5) |
All semi‐structured interviews were conducted by one to three members of the design team. All members of the interviewing team (B.S.A., D.P., L.C., and A.S.) had training in HCD approaches to semi‐structured interviewing through their extensive experience leading HCD research studies at an integrated HCD research lab (The Better Lab). Interview guides were collaboratively developed by the HCD‐trained members of the research team (B.S.A., D.P., L.C., and A.S.) and reviewed by all remaining co‐authors as well as the members of the Community Advisory Board. Interview guides are included in Appendix S1. Interviews were conducted in person, lasted approximately 30–60 min, and were audio‐recorded and transcribed. Field notes were also taken during interviews to capture additional insights from our conducted interviews.
2.4. Analysis
Interviews were analyzed using an inductive, HCD informed approach to thematic analysis. Researchers first summarized each interview on notecards, with each notecard representing a discrete observation or preliminary code, and transcribed key quotes as supporting evidence. Each interview was reviewed by at least two researchers, who then presented their summaries to the broader team for synthesis and discussion. These notecards were then organized onto large poster boards to facilitate visual comparison across interviews. Using a manual affinity‐mapping process—an HCD analysis/synthesis technique rooted in Grounded Theory—the research team collaboratively clustered related ideas and observations into emerging themes, visually grouping similar concepts to identify underlying patterns across interviews (C.N.S., B.A., D.P, L.C., and A.S.) [26, 27]. Coding and clustering were conducted manually to preserve the visual and participatory nature of the HCD analytic methods used (a technique commonly used in collective qualitative analysis in HCD studies). To strengthen analytic rigor, we used data triangulation by integrating findings across individual and group interviews and across stakeholder groups (clinical, community‐based, and patient participants), allowing the team to identify convergent and divergent perspectives among the respective groups. Themes were iteratively refined through team consensus meetings, with particular attention to tensions and alignment between stakeholders.
Once thematic coding was complete, researchers developed “insight statements” aiming to summarize underlying perspectives, motivations, and tensions within the thematically organized data. Based on these insight statements, “opportunity statements” were created to articulate specific, actionable “design opportunities” that addressed the challenges highlighted within each insight statement. After an initial analysis by four researchers (B.A., D.P., L.C., and A.S.), an additional synthesis session was held with the research team, Community Advisory Board members, and Solid Start leadership to refine and finalize the insight statements and iteratively develop the design opportunity statements until consensus was reached. This analytic process was consistent with established HCD methodology emphasizing visual, team‐based synthesis [26]. Consistent with HCD‐informed analytic approaches, the insight and design opportunity statements constitute the primary outputs of this study.
The University of California San Francisco's Institutional Review Board reviewed and exempted this study (#19‐27458). All participants provided verbal informed consent prior to participation, and patients were informed that their decision to participate was voluntary. No incentives or compensation were provided for participation. This study adhered to the Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines to ensure methodological transparency and rigor (see Table S1).
2.5. Researcher reflexivity
The research team comprised of a multidisciplinary group with clinical backgrounds spanning obstetrics and gynecology (C.N.S.), psychiatry (M.T.), pediatrics (N.D. and H.B.), surgery (A.S.), and public health nursing (M.S. and L.O.). It also included researchers with multiple years of experience leading HCD research studies in an HCD research laboratory specializing in design methodology (A.S., B.S.A., D.P., and L.C.). The inclusion of diverse clinical and design perspectives allowed for triangulation across professional orientations and enhanced the team's ability to interpret findings across patient, provider, and system levels.
All members of the analytic team were trained in qualitative and HCD methods. Clinical team members (C.N.S., M.T., N.D., H.B., M.S., L.O., and A.S.) drew on firsthand knowledge of inpatient and outpatient workflows, while design researchers (B.S.A., D.P., L.C., and A.S.) contributed expertise in participatory research and visual synthesis methods. Public health collaborators (M.S., and L.O.) grounded the analysis in the realities of community‐based perinatal care delivery within the local safety‐net system.
Throughout our analysis and synthesis, the team engaged in reflexive discussion regarding how their professional and personal identities might shape the interpretation of the data. Regular consensus meetings were used to critically examine these positional influences. The involvement of the Community Advisory Board in the synthesis process further served to ensure that the final insights and opportunity statements remained grounded in participant perspectives rather than researcher assumptions.
3. RESULTS
Through our analysis, we identified six insights and associated design opportunities to improve coordination at transitions of care for perinatal patients. These insights and associated design opportunities are described below. Table 2 provides a summary of our resulting insight statements and design opportunities with samples of supportive quotes from our semi‐structured interviews.
TABLE 2.
Summary of findings.
| Insight | Supporting quotes | Design opportunities |
|---|---|---|
| 1. The perinatal care system is resource‐rich, but poor coordination leads to fragmented care and inefficient resource utilization |
|
|
| 2. Safety‐net hospitals rely on community‐based organizations and providers, but lack reliable systems for communicating with them |
|
|
| 3. Due to patient data protections, the system default is to withhold information, which hinders collaborative care planning |
|
|
| 4. Poorly communicated substance use policies create uncertainty among providers, seeding bias into clinical decision‐making |
|
|
| 5. Patients value continuity of care, but the system's reliance on rotating providers across visits erodes trust in the care team |
|
|
| 6. The discharge process gives patients a false sense of security as they trust that follow‐up will be arranged, while providers who have limited control over coordination remain concerned about patients’ understanding. |
|
|
Abbreviation: COB, community‐based organization.
3.1. Insight 1: The perinatal care system is resource‐rich, but poor coordination leads to fragmented care and inefficient resource utilization
Although there are numerous resources available within the perinatal care system, coordination between them is often fragmented. While most services rely on referrals from prenatal care providers for recruitment, clinicians frequently have limited knowledge of the types of services that are available to their patients outside of their immediate clinics. As one physician stated, “Many groups attempt to create their own consolidated ‘resource lists’ to improve providers’ awareness of services…but each list is still separate, needs updating, and isn't universally accessed by perinatal providers.” This can lead to underutilization of certain services. For example, some specialty clinics in our safety‐net health system—such as a family medicine practice offering prenatal addiction care—are underutilized, not due to lack of capacity, but because they lack a structured referral pathway. Conversely, unclear delineation of service scope can lead to misutilization of more well‐known services. The obstetric psychiatry team within the safety‐net system, for instance, often receives referrals for full consultative visits when an electronic consultation for medication management would suffice. Both under‐ and misutilization hinder the efficient and effective use of available healthcare resources within the perinatal care system.
Poor communication between referring and receiving teams further contributes to fragmented care and diminishes the effectiveness of referrals. When referrals lack sufficient clinical context—such as the reason for referral, the specific concern to be addressed, or relevant patient history—receiving providers are often left to make assumptions or spend additional time gathering information. Patient reflected on the resulting experiences of disjointed or impersonal care due to poor coordination. As patient stated, “I have so many case managers from so many different agencies, none of it was centralized. For a while, a lot of them were like disconnected…nobody was communicating with each other… I was being referred to different agencies all over the place.” The breakdown in communication between teams not only limits the ability of receiving providers to offer timely, tailored care, but also contributes to inefficiencies across the system.
Design opportunities:
Develop a centralized, regularly updated system to clearly identify and communicate available perinatal resources, ensuring that providers can easily access and share relevant information.
Collaborate across prenatal care, labor and delivery, and pediatric settings to implement standardized referral pathways and communication protocols that promote coordinated perinatal care.
3.2. Insight 2: Safety‐net hospitals rely on CBOs and providers but lack reliable systems for communicating with them
CBOs and community‐based providers play a critical role in delivering comprehensive care to high‐risk perinatal patients yet are disconnected from the safety‐net system's communication systems. Connections between community‐based providers and clinical providers are inconsistent and in the rare instances in which they do occur are frequently an informal, one‐way communication from a clinical provider to a community‐based partner. As one public health nurse described, “information is shared through phone calls, emails, and sometimes in person… we don't have access to [the hospital system's] charting.” Relationships between individual CBOs and providers vary widely, and there is no standardized or reliable method for information exchange. Data systems used by community‐ and hospital‐based teams are incompatible, making it difficult to share updates about patient needs or progress (even with the permission from a patient to share information between hospital‐ and community‐based care teams). As a result, clinical providers are often unaware of the external services their patients are receiving or of important changes in patients’ social or economic circumstances.
Without timely information from community‐based partners, clinical providers are unable to integrate knowledge gained in community settings into their clinical care planning. This not only complicates the provision of clinical care, but ultimately may limit the impact and effectiveness of community‐based services. Patients experienced this lack of coordination in their perinatal care. One patient described taking ownership over sharing information between their clinical and community‐based care teams, stating, “Doctors don't talk to each other. It's okay because ‘I'm on top of it.’ But if I wasn't, and if I didn't have a case manager it would be a problem.” Ultimately, the lack of integration between community and clinical systems results in the provision of fragmented and siloes care, rather than sustainable and coordinated services.
Design opportunity:
Implement closed‐loop communication systems that facilitate timely, bidirectional exchange of information between community‐based and hospital‐based providers, ensuring that both social and clinical needs are identified, addressed, and continuously coordinated.
3.3. Insight 3: Due to patient data protections, the system default is to withhold information, which hinders collaborative care planning
Providers and staff who work with patients around sensitive issues—such as substance use or mental health—frequently express concerns about maintaining patient confidentiality. Concerns about maintaining patient privacy can lead to hesitancy among clinical team members to communicate or share pertinent information, hindering coordination of care. Additionally, confusion about the type of permission or consent required to share sensitive information in the service of clinical care is common. Providers described that although well‐known patient protection policies provide a general framework for protecting patient health information, certain topics—particularly information related to substance use disorder treatment—may be governed by additional federal regulations that require explicit patient authorization [28]. Providers suggested that these overlapping protections can contribute to clinician uncertainty and reluctance to share information, hindering the delivery of integrated care.
In addition, patients themselves often have individual preferences regarding who on their care team should have access to sensitive information. Some expect full transparency among all providers, while others prefer more selective sharing. Heterogeneity in patient preferences results in a default of not sharing. As one community‐based provider stated, “We're not reporting things to [the obstetricians]. That is up to the client, and if that's what you want to do the client needs to release that information via a [release of information].” Patients have their own needs and preferences for who on their care team has their information—to them, certain providers should know their whole story, and others should receive information as needed. This tension between building patient rapport and trust through confidentiality and the benefit of proactive information sharing complicates care coordination for some of the health system's highest‐risk patients.
Design opportunities:
Develop and implement standardized policies for information sharing between clinical teams and community‐based organizations to support coordinated care, ensuring that all stakeholders understand and apply these protocols consistently.
Respect patient autonomy in information disclosure by clearly documenting confidentiality preferences and streamlining data sharing when consent is provided. During care transitions, initiate structured discussions to align information sharing with patent wishes and clinical needs, bringing in appropriate members of the care team.
3.4. Insight 4: Poorly communicated substance use policies create uncertainty among providers, seeding bias into clinical decision‐making
Participants described substantial uncertainty among providers regarding when and how to apply institutional policies related to substance use during the perinatal period, particularly in decisions involving Child Protective Services (CPS) referrals and urine toxicology screening. Many providers were unaware that formal guidelines existed, while others described the policies as difficult to interpret due to ambiguous or inconsistent language. Frequent changes to the existing policies compounded this uncertainty for clinical providers. A pediatrics resident reflected on the challenge of staying current with institutional guidance around toxicology testing of substance‐exposed newborns, stating, “I have to consult the policy each time this occurs, as it frequently changes, and I need to ensure accuracy in my decision‐making.”
The lack of clarity and consistent communication regarding substance use protocols was perceived to drive conservative clinical behavior, in which providers err on the side of heightened surveillance or reporting. Participants expressed concern that, in the absence of clearly defined and consistently implemented guidelines, decisions may be informed by subjective judgment rather than objective criteria. For instance, a Labor and Delivery nurse described the decision‐making process surrounding calling CPS for parental substance use, stating, “Implementation is variable—there are no clear guidelines [and a] risk of bias as a result.” Similarly, a midwife reflected on how urine toxicology screening decisions can be influenced by subjective factors, noting, “In the event of ‘red flags’ like missing visits, altered behavior, [or if a patient] presents with trauma, ordering a [urine toxicology] for the mother is a realm that is especially susceptible to bias.”
Importantly, participants emphasized that the consequences of substance use policy ambiguity extend beyond provider behavior. Both providers and patients reported that inconsistent messaging about institutional policies contributed to patient confusion, mistrust, and the perception that care was punitive rather than supportive—further straining the therapeutic relationship between patients and providers during a highly vulnerable period.
Design opportunity:
Establish and regularly update clear, evidence‐based policies and clinical protocols for substance use–related care and implement a reliable system to ensure consistent dissemination and access for all staff, trainees, and providers.
3.5. Insight 5: Patients value continuity of care, but the system's reliance on rotating providers across visits erodes trust in the care team
Patients often enter perinatal care with the expectation of longitudinal relationships with consistent providers; however, in practice, care is frequently fragmented across multiple team members, which can undermine trust and continuity. As one patient described, “I saw many different people here…it's very different here…I'm used to seeing one person and that's it. But here you meet a bunch of different people.” Participants emphasized that while patients often come to their first prenatal appointment expecting to meet a provider who will follow them throughout pregnancy, they often undergo multiple intake appointments before they are able to establish care with a consistent provider. Patients who receive their care in teaching clinics are particularly susceptible to fragmentation of their care, as resident physicians rotate through clinics sporadically and rarely see patients twice. Similarly, when establishing pediatric care, new parents expect to meet with a physician who will provide longitudinal care for their child. Instead, patients with complex social needs are frequently referred to a social worker prior to meeting their pediatrician for the first time. Providers emphasized that when patients meet unfamiliar and unexpected providers, this can contribute to their distrust of the healthcare system. As one obstetrics resident physician described, “When patients meet unfamiliar and unexpected providers, it can contribute to their distrust of the healthcare system and limit their willingness to engage in care.” Although patients and providers discussed the negative consequences of shifting care providers, their discussions of potential solutions centered on improved communication and expectation setting, rather than proposed solutions to restructure care provision to facilitate provider continuity.
Design opportunities:
Comprehensively inform and onboard patients to the variety of perinatal care services and types of providers that they can expect to interact with over the course of their perinatal journey.
Implement structured warm handoffs and facilitate intentional, relationship‐centered introductions when transitioning patients to new providers or services to support trust‐building and care continuity.
3.6. Insight 6: The discharge process gives patients a false sense of security as they trust that follow‐up will be arranged, while providers who have limited control over coordination remain concerned about patients’ understanding
Many providers expressed concern that patients may be discharged from the hospital without a clear understanding of when and where to complete their follow‐up care. In the postpartum setting, for example, the plan at discharge for scheduling a follow‐up appointment is typically either for the clinic to call the patient after they leave the hospital or for the patient to call the clinic to establish care. In the studied healthcare system, it is rare for follow‐up appointments to be confirmed prior to the patient leaving the hospital. Despite this, patients commonly assumed that their providers would take care of arranging all their follow‐up. One patient preparing for same‐day discharge explained, “I assume they'll just schedule me. I'll go wherever they say,” while another shared, “I don't have an appointment, but the nurse will give one to me. The doctors will let me know [about] next steps. Everything else is common sense. Motherly instinct.” Patients frequently expressed trust that transitions in care would be taken care of by the inpatient care team.
Contrastingly, providers described this hospital discharge process as creating a false sense of security for patients. They explained that when follow‐up plans are not clearly documented or confirmed before discharge, patients often leave the hospital without realizing they may need to take the initiative to arrange or navigate their next steps in care. Providers expressed that discharging patients without confirmed appointments increases the risk of delayed or missed follow‐up. As one social worker described, “I track the no‐shows after clinic. Before me, the women just disappeared—they had an appointment, no‐showed, and nobody followed up.” This tension—between patients’ confidence that their care would be coordinated and the reality of discharge practices that rely on patient initiative—highlighted for providers the need for more structured and transparent care transition processes, particularly for populations managing the complexities of the prenatal and postpartum periods.
Design opportunities:
Develop systems that enable inpatient providers to schedule follow‐up appointments prior to discharge, reducing ambiguity in the follow‐up plan.
Facilitate dedicated discharge conversations that clearly summarize next steps, allow patients to ask questions, and clarify their understanding of the follow‐up plan.
4. DISCUSSION
This study used an HCD approach to identify opportunities to improve care coordination and optimize transitions of care for perinatal patients at an urban safety‐net hospital. We identified key challenges within the current perinatal care ecosystem, including siloed care, fragmented documentation, and limited communication between community‐ and hospital‐based stakeholders. In addition to creating clinical and logistical challenges for providers, fragmented communication eroded patients’ trust in the healthcare system. Our HCD process revealed key opportunities to improve care coordination. Interventions to strengthen informational continuity between clinical providers, community‐based organizations, and families are promising opportunities to improve the delivery of perinatal care in safety‐net systems.
The findings presented in this study build on an emerging literature examining transitions of care and care coordination during the perinatal period. In perinatal contexts, transitions between prenatal, intrapartum, and postpartum care are particularly vulnerable to breakdowns in continuity of care, as patients shift between inpatient and outpatient care teams. This is especially true for those with complex social needs, who receive care from multiple clinical and community‐based providers who often lack communication systems with each other [29]. Participants in our study described a perinatal ecosystem that was resource‐rich but poorly coordinated. While many clinical programs and community‐based organizations existed to support pregnant and parenting people, siloed systems resulted in missed opportunities to connect patients to available services. Our HCD process revealed that despite the availability of numerous services to support patients with complex social needs, providers were unaware of the scope of existing clinical teams and organizations. Various studies have demonstrated that coordinated care leads to improved outcomes for publicly insured parents and newborns, likely through higher rates of enrollment in specialty services [1, 19, 21]. Our solution‐focused design work suggested that optimizing referrals within the existing perinatal ecosystem, rather than focusing exclusively on adding new resources, is an important step toward improved patient care.
While sharing patient information is essential for safe transitions of care, our study highlighted challenges of inter‐provider communication within our safety‐net system. In the United States, patient health records are frequently spread across multiple electronic record systems with minimal interoperability. Publicly insured patients served at safety‐net hospitals are more likely to receive care from multiple institutions, compounding this challenge [29]. Patients, clinical providers, and community organizations in our study described the challenges of providing high‐quality care when information cannot be readily shared between care teams. Providers must have a deep understanding of patients’ needs and existing service connections to facilitate meaningful referrals, and lost information inhibits their ability to connect patients to relevant resources. Continuity of care has been linked to patient confidence in their providers [30], and patients in our study reported frustration with the lack of communication between their care teams. Studies across perinatal settings have shown that lapses in informational continuity during transitions of care contribute to delays in follow‐up and redundant care processes, and lead to patient mistrust—themes that were reflected in our study [12, 14]. Our results suggest that dynamic communication systems that allow for information to flow between community‐based and clinical systems offer the potential to improve informational continuity between community‐ and hospital‐based providers, improving patient care experiences.
Consistent policies around using sensitive topics, such as substance use disorders, are vital to maintaining patient trust. Our study revealed that policies for calling child welfare or ordering urine toxicology screenings for patients with substance use disorder within our ecosystem were ambiguous, leaving space for bias in patient treatment. Racial disparities in both toxicology screenings [31] and child welfare referrals [32] are well‐documented in the literature and lead to lasting family trauma [33, 34]. Fear of child separation contributes to “system” avoidance, resulting in decreased use of social services and the strategic concealment of information from providers and staff with access to resource referral [35]. Our study highlights how divergent understandings of policies and procedures cause confusion for providers and risk the introduction of bias into clinical decision‐making. Our results suggested that standardized policies that minimize opportunities for bias and clear communication between all providers and staff may be a first step toward creating more equitable care environments. Restoring trust in our care institutions will require significant patient outreach and a system‐wide recommitment to building equitable and patient‐centered environments.
This study represents the first application of HCD methodologies to examine the processes of care transition and coordination within a complex perinatal care ecosystem. By employing HCD in this context, we contribute both a novel perspective and a methodological approach uniquely suited to elucidate the challenges of providing care for pregnant and postpartum patients in safety‐net settings. The use of HCD methodologies in perinatal care systems research remains limited. Nijagal et al. applied HCD methods to identify system‐level barriers contributing to inequities in perinatal care among Medicaid‐insured patients, highlighting the value of design approaches in safety‐net contexts [27]. Similarly, other authors have used HCD for specific system improvement efforts in the perinatal space, such as redesigning social determinants of health screening workflows [36] or improving the physical facilities of Labor and Delivery units [37]. In a recent review of HCD studies in the perinatal care space, Landeiro et al. note the limited utilization of HCD methodologies in the perinatal research and highlight the value of participatory and ethnographic methods to uncover user‐level perspectives within complex systems of care [38]. Our study contributes to the emerging body of literature surrounding transitions of care and coordination for perinatal patients by uniquely applying HCD methodologies to a complex, safety‐net ecosystem. In using an HCD approach, we illustrate how participatory methods can generate contextually grounded insights that ultimately inform system‐level redesign for complex patient care systems.
Various limitations impacted the results of this study. First, while we followed the HCD process in generating opportunity statements from our insights, these opportunities have not been tested or assessed. In the next phase of this work, a deeper assessment of the feasibility of our identified opportunities will be conducted, which may subsequently lead to the reframing of both opportunities and insights. In addition, while interview participants were selected to increase the diversity of data obtained and produce rich and actionable insights, this strategy may have introduced bias in the selection of study participants. In addition, this study has a limited sample size. While smaller sample sizes are both inherent to and necessary for rapid iteration in HCD, the generalizability of our findings may be more limited as a result. The HCD methodologies that we adopted facilitated the engagement of a wide range of stakeholders in a complex care system.
Finally, although participants emphasized the importance of continuity of care with an established provider to build trusting patient–provider relationships, few proposed structural changes to clinic staffing or scheduling models that would enable patients to consistently see the same provider. Instead, participants focused on solutions centered on improving communication, transparency, and coordination within the existing care structure (e.g., through standardized handoffs and shared care‐team documentation). In general, participants discussed opportunities for low‐barrier, system‐level improvements to strengthen communication and trust without requiring major organizational redesign. The limited discussion of broader structural reforms highlights a missed opportunity to envision new models of care that could reduce handoffs and better support longitudinal continuity across the perinatal period.
This study identified key challenges and high‐impact areas in which interventions to improve care coordination for perinatal patients may be particularly successful. The design opportunity statements generated through our HCD process will serve as the foundation for the development and implementation of prototype solutions in the next phases of this work. Improving care coordination across the perinatal ecosystem will require efforts to strengthen informational continuity, especially around transitions of care. Optimizing the utilization of existing resources, rather than generating redundant services, is a more sustainable approach to increasing access to vital services for some of our highest‐risk patients.
CONFLICT OF INTEREST STATEMENT
Amanda Sammann, Devika Patel, and Lara Chehab serve as design consultants for The Empathy Studio. All other authors declare no conflicts of interest.
ETHICS STATEMENT
This study was approved by the Institutional Review Board of the University of California San Francisco.
Supporting information
Supporting Information
Supporting Information
ACKNOWLEDGMENTS
We thank the members of the Plan of Safe Care Community Collaborative and Solid Start for their guidance on study design and interpretation. We also appreciate Schyneida Williams and Saeed Mifattah for their contributions to reviewing insights and design opportunities. This work was supported by funding from the San Francisco General Hospital Foundation.
DATA AVAILABILITY STATEMENT
Qualitative data supporting the findings of this study may be available upon reasonable request.
REFERENCES
- 1. Kroll‐Desrosiers, A. R. , Crawford S. L., Moore Simas T. A., Rosen A. K., and Mattocks K. M.. 2016. “Improving Pregnancy Outcomes Through Maternity Care Coordination: A Systematic Review.” Women's Health Issues 26(1): 87–99. 10.1016/j.whi.2015.10.003. [DOI] [PubMed] [Google Scholar]
- 2. Heaman, M. I. , M. Moffatt, , Elliott L., Sword W., Helewa M. E., Morris H., Gregory P., Tjaden L., and Cook C., 2014. “Barriers, Motivators and Facilitators Related to Prenatal Care Utilization Among Inner‐City Women in Winnipeg, Canada: A Case‐Control Study.” BMC Pregnancy and Childbirth 14(1): 227. 10.1186/1471-2393-14-227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Mazul, M. C. , Salm Ward T. C., and Ngui E. M.. 2017. “Anatomy of Good Prenatal Care: Perspectives of Low Income African‐American Women on Barriers and Facilitators to Prenatal Care.” Journal of Racial and Ethnic Health Disparities 4(1): 79–86. 10.1007/s40615-015-0204-x. [DOI] [PubMed] [Google Scholar]
- 4. Paisi, M. , March‐McDonald J., Burns L., Snelgrove‐Clarke E., Withers L., and Shawe J.. 2021. “Perceived Barriers and Facilitators to Accessing and Utilising Sexual and Reproductive Healthcare for People Who Experience Homelessness: A Systematic Review.” BMJ Sexual & Reproductive Health 47(3): 211–20. 10.1136/bmjsrh-2020-200799. [DOI] [PubMed] [Google Scholar]
- 5. Gelberg, L. , Browner C. H., Lejano E., and Arangua L.. 2004. “Access to Women's Health Care: A Qualitative Study of Barriers Perceived by Homeless Women.” Women & Health 40(2): 87–100. 10.1300/J013v40n02_06. [DOI] [PubMed] [Google Scholar]
- 6. Lennox, R. , Patel T., Marmel A., and Shaw E.. 2021. “Prenatal Care Outcomes in Women With Substance Use Disorders: A Retrospective Cohort Study.” Journal of Obstetrics and Gynaecology Canada 43(7): 850–5. 10.1016/j.jogc.2020.10.021. [DOI] [PubMed] [Google Scholar]
- 7. Weir, S. , Posner H. E., Zhang J., Willis G., Baxter J. D., and Clark R. E.. 2011. “Predictors of Prenatal and Postpartum Care Adequacy in a Medicaid Managed Care Population.” Women's Health Issues 21(4): 277–85. 10.1016/j.whi.2011.03.001. [DOI] [PubMed] [Google Scholar]
- 8. Testa, A. , and Jackson D. B.. 2020. “Incarceration Exposure and Barriers to Prenatal Care in the United States: Findings From the Pregnancy Risk Assessment Monitoring System.” International Journal of Environmental Research and Public Health 17(19): E7331. 10.3390/ijerph17197331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Cha, S. , and Masho S. W.. 2014. “Intimate Partner Violence and Utilization of Prenatal Care in the United States.” Journal of Interpersonal Violence 29(5): 911–27. 10.1177/0886260513505711. [DOI] [PubMed] [Google Scholar]
- 10. Alhusen, J. L. , Ray E., Sharps P., and Bullock L.. 2015. “Intimate Partner Violence During Pregnancy: Maternal and Neonatal Outcomes.” Journal of Women's Health 24(1): 100–6. 10.1089/jwh.2014.4872. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Krans, E. E. , Davis M. M., and Palladino C. L.. 2013. “Disparate Patterns of Prenatal Care Utilization Stratified by Medical and Psychosocial Risk.” Maternal and Child Health Journal 17(4): 639–45. 10.1007/s10995-012-1040-9. [DOI] [PubMed] [Google Scholar]
- 12. Saultz, J. W. 2003. “Defining and Measuring Interpersonal Continuity of Care.” Annals of Family Medicine 1(3): 134–43. 10.1370/afm.23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Haggerty, J. L. , Roberge D., Freeman G. K., and Beaulieu C.. 2013. “Experienced Continuity of Care When Patients See Multiple Clinicians: A Qualitative Metasummary.” Annals of Family Medicine 11(3): 262–71. 10.1370/afm.1499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Haggerty, J. L. , Reid R. J., Freeman G. K., Starfield B. H., Adair C. E., and McKendry R.. 2003. “Continuity of Care: A Multidisciplinary Review.” BMJ 327(7425): 1219–21. 10.1136/bmj.327.7425.1219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Wong, P. C. , and Kitsantas P.. 2020. “A Review of Maternal Mortality and Quality of Care in the USA.” Journal of Maternal‐Fetal & Neonatal Medicine 33(19): 3355–67. 10.1080/14767058.2019.1571032. [DOI] [PubMed] [Google Scholar]
- 16. Patient Safety and Quality Committee, Society for Maternal‐Fetal Medicine ; Gibson, K. S. , and McLean D.. 2020. “Society for Maternal‐Fetal Medicine Special Statement: A Maternal Transport Briefing Form and Checklist.” American Journal of Obstetrics and Gynecology 223(5): B12–5. 10.1016/j.ajog.2020.08.065. [DOI] [PubMed] [Google Scholar]
- 17. O'Rourke, K. , Teel J., Nicholls E., Lee D. D., Colwill A. C., and Srinivas S. K.. 2018. “Improving Staff Communication and Transitions of Care Between Obstetric Triage and Labor and Delivery.” Journal of Obstetric, Gynecologic, and Neonatal Nursing 47(2): 264–72. 10.1016/j.jogn.2017.11.008. [DOI] [PubMed] [Google Scholar]
- 18. Centers for Medicare and Medicaid Services . Resources on Strategies to Improve Postpartum Care among Medicaid and CHIP Populations. Published online February 2015.
- 19. Willems Van Dijk, J. A. , Anderko L., and Stetzer F.. 2011. “The Impact of Prenatal Care Coordination on Birth Outcomes.” Journal of Obstetric, Gynecologic, and Neonatal Nursing 40(1): 98–108. 10.1111/j.1552-6909.2010.01206.x. [DOI] [PubMed] [Google Scholar]
- 20. Buescher, P. A. , Roth M. S., Williams D., and Goforth C. M.. 1991. “An Evaluation of the Impact of Maternity Care Coordination on Medicaid Birth Outcomes in North Carolina.” American Journal of Public Health 81(12): 1625–9. 10.2105/ajph.81.12.1625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Hillemeier, M. M. , Domino M. E., Wells R., Goyal R. K., Kum H.‐C., Cilenti D., and Basu A.. 2018. “Does Maternity Care Coordination Influence Perinatal Health Care Utilization? Evidence from North Carolina.” Health Services Research 53(4): 2368–83. 10.1111/1475-6773.12742. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Brown, T. 2008. “Design Thinking.” Harvard Business Review 86(6):84–92, 141. https://hbr.org/2008/06/design‐thinking. [PubMed] [Google Scholar]
- 23. Ideo . Design Thinking. Accessed December 15, 2021. https://designthinking.ideo.com/.
- 24. Ideo . The Field Guide to Human Centered Design. Accessed December 15, 2021. https://www.designkit.org/resources/1.
- 25. Proctor, E. , Silmere H., Raghavan R., Hovmand P., Aarons G., Bunger A., Griffey R., and Hensley M.. 2011. “Outcomes for Implementation Research: Conceptual Distinctions, Measurement Challenges, and Research Agenda.” Administration and Policy in Mental Health 38(2): 65–76. 10.1007/s10488-010-0319-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Simonsen, J. , and Friberg K.. 2014. “Collective Analysis of Qualitative Data.” In Situated Design Methods. Design Thinking, Design Theory, edited by Simonsen J., Svabo C., Strandvad S. M., Samson K., Hertzum M., and Hansen O. E., 99–117. MIT Press; Accessed October 16, 2025. http://mitpress.mit.edu/books/situated‐design‐methods. [Google Scholar]
- 27. Nijagal, M. A. , Patel D., Lyles C., Liao J., Chehab L., Williams S., and Sammann A.. 2021. “Using Human Centered Design to Identify Opportunities for Reducing Inequities in Perinatal Care.” BMC Health Services Research 21(1): 714. 10.1186/s12913-021-06609-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. U.S. Department of Health and Human Services . 2020. Confidentiality of Substance Use Disorder Patient Records. Accessed June 13, 2025. https://www.ecfr.gov/current/title‐42/part‐2. [Google Scholar]
- 29. Sorace, J. , Wong H.‐H., DeLeire T., Xu D., Handler S., Garcia B., and MaCurdy T.. 2020. “Quantifying the Competitiveness of the Electronic Health Record Market and Its Implications for Interoperability.” International Journal of Medical Informatics 136: 104037. 10.1016/j.ijmedinf.2019.104037. [DOI] [PubMed] [Google Scholar]
- 30. Nowak, D. A. , Sheikhan N. Y., Naidu S. C., Kuluski K., and Upshur R. E. G.. 2021. “Why Does Continuity of Care With Family Doctors Matter? Review and Qualitative Synthesis of Patient and Physician Perspectives.” Canadian Family Physician 67(9): 679–88. 10.46747/cfp.6709679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Perlman, N. C. , Cantonwine D. E., and Smith N. A.. 2022. “Racial Differences in Indications for Obstetrical Toxicology Testing and Relationship of Indications to Test Results.” American Journal of Obstetrics & Gynecology MFM 4(1): 100453. 10.1016/j.ajogmf.2021.100453. [DOI] [PubMed] [Google Scholar]
- 32. Putnam‐Hornstein, E. , Ahn E., Prindle J., Magruder J., Webster D., and Wildeman C.. 2021. “Cumulative Rates of Child Protection Involvement and Terminations of Parental Rights in a California Birth Cohort, 1999–2017.” American Journal of Public Health 111(6): 1157–63. Published online April 15, 2021:e1‐e7. 10.2105/AJPH.2021.306214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Merritt, D. H. 2021. “Lived Experiences of Racism Among Child Welfare‐Involved Parents.” Race and Social Problems 13(1): 63–72. 10.1007/s12552-021-09316-5. [DOI] [Google Scholar]
- 34. Merritt, D. H. 2020. “How Do Families Experience and Interact with CPS?.” ANNALS of the American Academy of Political and Social Science 692(1): 203–26. 10.1177/0002716220979520. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Fong, K. 2019. “Concealment and Constraint: Child Protective Services Fears and Poor Mothers' Institutional Engagement.” Social Forces 97(4): 1785–10. 10.1093/sf/soy093. [DOI] [Google Scholar]
- 36. Umstead, K. A. , Gill C., Pearsall M. S., Stuebe A. M., and Tully K. P.. 2023. “Human‐Centered Design in the Context of Social Determinants of Health in Maternity Care: Methods for Meaningful Stakeholder Engagement.” International Journal of Qualitative Studies on Health and Well‐being 18(1): 2205282. 10.1080/17482631.2023.2205282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Sherman, J. P. , Hedli L. C., Kristensen‐Cabrera A. I., Lipman S. S., Schwandt D., Lee H. C., Sie L., et al. 2020. “Understanding the Heterogeneity of Labor and Delivery Units: Using Design Thinking Methodology to Assess Environmental Factors That Contribute to Safety in Childbirth.” American Journal of Perinatology 37(6): 638–46. 10.1055/s-0039-1685494. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Landeiro, F. , Silva M., Moura C. V. E., Martins C., Miller P., Ferraz S., de Azevedo A. P., et al. 2025. “Human‐Centered Design and Maternity Care: Is This a Possible Interplay?—A Systematic Review.” BMC Pregnancy and Childbirth 25(1): 261. 10.1186/s12884-024-07119-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information
Supporting Information
Data Availability Statement
Qualitative data supporting the findings of this study may be available upon reasonable request.