Adapting a complex health intervention guided by the core functions and forms model: a methodology and case example of participatory adaptation Authors: Kirsten E. Austad, Nhi Nguyen, Elizabeth Jadovich, Bridget Poznanski, Max Laguerre, Alicia Fernandez, Mari-Lynn Drainoni, Brian Jack, Suzanne Mitchell, Ramzi G. Salloum, Graham Moore Published: 2026 ABSTRACT Intervention adaptation—the intentional modification of a program, practice, or innovation's design or delivery—is essential for improving fit when interventions are implemented with diverse populations, yet how to adapt without compromising core mechanisms remains a persistent challenge for complex, multicomponent programs. This article describes how the Core Functions and Forms (CFF) model, used in conjunction with the ADAPT guidance, addresses this challenge by distinguishing essential intervention mechanisms (core functions) from adaptable delivery strategies (forms). To illustrate the approach, we present a case example of adapting hospital discharge interventions for patients with non-English language preference. A 13-member Adaptation Team of patients, caregivers, and hospital staff participated in four facilitated sessions to identify and prioritize core functions shared across existing discharge interventions, then propose context-appropriate forms—distinguishing specific forms tied to individual functions from cross-cutting forms spanning who delivers the intervention, when, where, and how language concordance is achieved. Rankings and impact–feasibility ratings informed deliberative prioritization, surfacing functions that balanced high impact with feasibility as well as those judged impactful but harder to implement. Applying the CFF model across a family of related interventions, rather than a single program, shows how shared mechanisms can be reused and tailored for diverse populations. This manuscript reports the adaptation-design phase; subsequent pilot testing is planned. The approach offers a transparent, replicable method for equity-focused adaptation that preserves intervention mechanisms while tailoring delivery. Introduction Intervention adaptation, the intentional modification of a program, practice, or innovation's design or delivery, is a common strategy to improve an intervention's fit within specific contexts (1). When evidence-based interventions (EBIs) are implemented with diverse populations, adaptation plays a crucial role in aligning interventions with the unique needs and preferences of these groups (2–4). This alignment can enhance an intervention's implementability and effectiveness and promote health equity, making adaptation a vital component of efforts to achieve equitable health outcomes (5–8). Despite its importance, many questions remain about the best approaches to adapting interventions, particularly in high diversity contexts (9, 10). Ensuring an EBI fits the local context is a widely accepted first step in implementation. Approaches such as ADAPT and ADAPT-ITT explicitly require assessing intervention-context fit and recommend that adaptations remain aligned with the intervention's functions (11–13). A persistent tension is tailoring interventions while ensuring that effect-producing elements are preserved. Frameworks such as Consolidated Framework for Implementation Research (CFIR) distinguish “core components” vs. “adaptable periphery” (14–18). However, in practice it is difficult to specify which elements are truly essential or to predict which adaptations will alter an intervention's mechanism of action or have unintended effects (19–21). This uncertainty is greater for complex interventions, those that feature multiple interacting components, variable implementation processes, heterogeneous outcomes, and complex interactions with context, because it is harder to measure and track which elements are responsible for the observed effects (22–24). One approach that may address these challenges is the Core Functions and Forms (CFF) model (25–27). The CFF operationalizes ideas first advanced by Hawe and colleagues in 2004 around defining intervention integrity in terms of functions, while allowing forms to vary across context (25). CFF posits that interventions are defined by a consistent set of core functions (the essential mechanisms of an intervention to preserve), while the forms (the specific ways these functions are delivered) can and should vary based on the needs of the implementation context and population. For example, consider an evidence-based patient education intervention originally developed and tested using an English-language handout. The core function of the EBI is “to help patients understand their illness.” If one sought to adapt the EBI for a setting where many patients prefer Spanish or have limited literacy, the same function would not be achieved by an English language handout but would require a written Spanish handout, illustrated materials, or a language-concordant audio recording of the content of the written materials. Not adapting the form—e.g., providing only English text—would in fact undermine its effectiveness. By centering adaptation on functions rather than fixed components, CFF offers a principled way to resolve the fidelity-tailoring paradox which aims to preserve the mechanisms that produces benefits (the function), while permitting flexible, context-appropriate changes in delivery (the form). A deeper understanding of these core functions and adaptable forms offers a useful conceptual model for equity-focused intervention adaptation. This manuscript aims to describe the application of the CFF model to support stakeholder-engaged intervention adaptation and analyze the benefits and drawbacks of this approach. We focus on a case study involving the adaptation of a hospital discharge EBI for patients with a U.S. safety net hospital. Specifically, we applied the CFF model within the ADAPT process to define core functions and identify context-appropriate forms for hospital discharge interventions serving NELP populations (11). Case example: evidence-based intervention to improve hospital discharge The transition from hospital to home after hospital discharge is a high-risk and vulnerable period for both patients and caregivers (28). Hospital discharge processes require coordination among hospital teams, clear handoffs between inpatient and outpatient clinicians, accurate medication reconciliation, and instructions to patients and caregivers that are both understandable and actionable. Undelivered or unclear communication, ambiguous delineation of responsibilities, and unrealistic expectations placed on patients and their caregivers (29–31) are common and contribute to high readmission rates and low patient satisfaction (32). Several hospital discharge interventions have been shown to reduce hospital readmission rates and improve patient-centered outcomes, such as Project RED and the Care Transitions Intervention (33–38),. These hospital discharge EBIs exemplify complex interventions because they include multiple interacting components, generate non-linear causal pathways, vary substantially in how they are implemented across settings, and are contextually dependent (39). An important limitation, however, is that nearly all hospital discharge EBIs have been designed for and tested only in English-fluent populations (40, 41). Individuals with non-English language preference (NELP), a growing demographic in the United States over the past 40 years (42), have been routinely excluded from hospital discharge trials, yet are at very high risk for poor outcomes due to inherent communication barriers (43, 44). As a result, these EBIs do not address language-related barriers that arise in routine practice, such as translating provider-written discharge instructions, overcoming language barriers in verbal communication, or consider other unique needs of those with NELP (45, 46). Without deliberate adaptation, deploying these EBIs in real-world, linguistically diverse settings risks reducing their effectiveness and widening health inequities (47–49). Because discharge EBIs are complex, a function-based approach is particularly useful for ensuring essential mechanisms remain intact when adapting for linguistically diverse populations. Methods Overview of approach The overall goal of this study was to adapt a hospital discharge intervention for patients with NELP for future pilot testing. Our assessment of intervention-context fit (discussed above) found that no single existing hospital discharge intervention would fit the needs of populations with NELP. However, many hospital discharge EBIs share common components, such as patient education, medication reconciliation, care coordination, and post-discharge outreach (38, 39). Rather than adapting a single EBI, we therefore framed adaptation around core functions (what an intervention must achieve) rather than forms (specific program components), aiming to preserve the mechanisms that produce benefit while tailoring delivery to the needs of patients with NELP. The author team selected the ADAPT guidance to guide the process, with tasks divided between the author group (italicized steps in Figure 1) and the Adaptation Team (non-italicized steps) (11). Step-by-step depiction of how the ADAPT guidance and the core functions and forms (CFF) model were integrated to adapt a hospital discharge intervention for patients with non-English language preference (NELP). Steps of the ADAPT guidance are shown left to right; content of each Adaptation Team meeting is shown top to bottom. Light blue boxes denote steps that focused on core functions and dark blue boxes those that focused on intervention forms. Tasks shown in italics were led by the study team; those in regular type were led by the Adaptation Team. How the core functions and forms (CFF) model was applied to adapt a hospital discharge intervention for patients with non-English language preference (NELP). The research team first established an initial list of candidate core functions through evidence synthesis. Drawing on a recent systematic review (39), we identified hospital discharge interventions conducted in the United States (excluding those limited to pediatric, psychiatric, or surgical admissions) (33, 50–82). The lead author reviewed the primary publication for each included intervention to abstract its core functions. The Adaptation Team was given three tasks: (1) review the draft list of core functions from literature review, add core functions important for patients with NELP, and prioritize them; (2) propose context-specific adaptable forms (the activities, materials, and delivery strategies that could reliably achieve each core function in a multilingual, multicultural care context); and (3) provide iterative feedback on proposed functions and forms to inform pilot implementation. Forms were of two types: specific forms correspond to a single core function—the concrete activity, material, or strategy used to achieve it—while cross-cutting forms operationalize multiple core functions at once: who delivers the intervention, and where, when, and how it is delivered (notably, how language barriers are addressed Figure 2). Study setting Study activities occurred at Boston Medical Center (BMC), the largest safety net hospital network in New England. BMC serves a diverse population, with the majority of patients identifying as Black or Latinx. Approximately one-third of all patients have NELP. This high linguistic diversity context includes a patient population that is about 54% Spanish speaking, 18% Haitian Creole speaking, 9% Cape Verdean (Portuguese) Creole speaking, 7% Vietnamese speaking, and the remainder speaking over 250 additional languages. Participant recruitment We aimed to recruit 10–14 Adaptation Team members, including patient and caregiver representatives and BMC employees. Patient and caregiver representatives were eligible if they had direct experience with hospital discharge as a patient with NELP or had assisted a patient with NELP during discharge. We prioritized recruiting individuals whose first language was not English and advertised the availability of simultaneous interpretation and translation services to enable participation by monolingual individuals with NELP. To identify these representatives, we engaged community networks and used snowball sampling to reach individuals outside BMC with relevant lived experience. Non-employee participants received $125 pre-loaded debit cards for each adaptation meeting attended and for completing voting and survey activities. We sought to include BMC employees across provider types involved in hospital discharge, including physicians, advanced practice providers, nurses, case managers, and social workers. To recruit BMC staff, we emailed supervisors in nursing, interpreter services, social work, and inpatient provider departments, who then shared study information with potentially interested team members. Adaptation team meetings We convened the Adaptation Team via videoconference four times between May and June 2025, with each 90-minute session facilitated by the PI. To encourage participation, the PI used slides alongside interactive Mentimeter polls for voting/ranking and comment boxes (83). Voting exercises served to surface opinions and stimulate discussion rather than to impose hard cutoffs, as in a formal Delphi process (84). Table 2 summarizes meeting topics. Each session began with a short recap and open discussion. Team members who missed a session were offered the opportunity to meet with the PI or review the written transcript. Meeting 1 opened with introductions and a process-mapping exercise to describe the hospital discharge process from an NELP patient's perspective, followed by a presentation of the literature review of discharge EBIs, and a group brainstorming of challenges, strengths, and implications for populations with NELP. In Meeting 2, we presented a revised process map and then introduced the CFF model, walking participants through an everyday example to illustrate the concepts and invite questions. To reinforce the distinction between core functions and forms, we used a consistent color convention on all slides. We then reviewed the initial list of candidate core functions derived from the literature, which team members critiqued and revised drawing on their clinical and lived experience. In Meeting 3 the group discussed their preliminary ranking and ratings of the impact and feasibility of core functions via Mentimeter polls and discussed emerging patterns. Next, they brainstormed intervention forms for each core function. For specific forms, participants brainstormed as many ideas as time allowed. The research team derived candidate cross-cutting forms from literature on commonly adapted aspects of EBIs (85); participants were asked to begin considering the feasibility and impact of each using Mentimeter polls, which laid out the groundwork for the formal voting and ranking process described below. Meeting 4 presented voting results, facilitated discussion, and solicited feedback on the proposed pilot model based on information collected in earlier meetings within the constraints of alignment with the funded grant. Data collection Demographic information for each Adaptation Team member was collected using a REDCap survey. All four meetings were audio-recorded and transcribed using Zoom's AI transcription feature. Transcripts were then reviewed by a research assistant (RA) for accuracy and identifying information was redacted to maintain participant confidentiality. Rating by all Adaptation Team members was conducted using REDCap surveys. Members ranked a list of 15 core functions of a hospital discharge intervention for patients with NELP on a scale from 1 to 15, where 1 indicated the core function with the highest importance and 15 the lowest. Additionally, for each core function, members assigned two scores from 1 to 10: one representing the perceived impact of the core function and the other representing the feasibility of incorporating it into a hospital discharge intervention. Lower scores indicated lower impact or feasibility, respectively. The Adaptation Team also rated the impact and feasibility of cross-cutting forms, Participants were asked to assign scores from 1 to 10 to indicate the perceived feasibility and impact across the following questions: where and when the intervention should be delivered, who should deliver the intervention, and how to overcome the language barrier. Data analysis Two analysts independently reviewed the transcripts from each meeting. We used conventional content analysis with descriptive coding to extract and enumerate discrete ideas or suggestions from participants. Prior to the next meeting, these ideas were compiled into a consolidated list, which served as the basis for subsequent discussion and rating. The research team then conducted summative analyses of the Adaptation Team's input on core functions. From the core function ranking exercise we calculated an average ranking to summarize the group's relative prioritization. Next, we computed mean impact and mean feasibility scores from the voting exercise. We produced a bubble plot to display these results with mean feasibility on the x-axis, mean impact on the y-axis, and one bubble per core function; bubble size represented standard deviation, so larger bubbles indicated greater variance in opinions. This visualization highlighted functions that combined high impact with high feasibility, as well as those judged high impact but less feasible to implement. We applied the same analytic approach to each of the crosscutting forms, generating analogous bubble plots to summarize perceived impact–feasibility tradeoffs for those elements. Lastly, we abstracted from transcripts all ideas for specific forms generated by the Adaptation Team and organized them by corresponding core function. Ethics This study was approved by the Boston University Medical Center Institutional Review Board (H-44249), including a waiver of documentation of written consent. All participants provided verbal informed consent at the start of each Adaptation Team meeting. Results Adaptation team and meeting structure In total, 13 individuals participated in the Adaptation Team. Participant demographics are provided in Table 1. All participants were fluent in English, 69% spoke at least one additional language and 54% spoke a language other than English at home. Most team members (77%), including non-BMC employees, were employed in health-care related fields. The team was made up of primarily females (92%) who were young adults 35 years or under (69%) and had previously or currently acted as caregiver for a loved one (85%). Table 1Demographics of adaptation team members.DemographicN (%)Gender Male1 (8) Female12 (92)Age 18–24 years3 (23) 25–35 years6 (46) 36–45 years3 (23) 46–55 years0 (0) 55 years or older1 (8)Profession Health care worker: physician, advanced practice provider, or resident4 (31) Health care worker: Interpreter2 (15) Health care worker: Social worker3 (23) Health care worker: Other1 (8) Other3 (23)Languages of fluency English13 (100) Spanish5 (38) Haitian Creole3 (23) Vietnamese1 (8) French1 (8)Language spoken most often at home English6 (46) Spanish3 (23) Haitian Creole2 (15) Vietnamese2 (15)Ever been hospitalized Yes8 (62) No5 (38)Ever acted as a caregiver Yes11 (85) No2 (15)Boston Medical Center employee Yes7 (54) No6 (46) Demographics of adaptation team members. Initially we planned to conduct voting during Meetings 2 and 3 using Mentimeter. However, based on Adaptation Team feedback requesting more time to deliberate—both on the distinction between core functions and forms and on ranking choices—we moved to asynchronous polling after Meeting 3, distributing polls via email (Table 2). Attendance was relatively consistent throughout the course of the adaptation process with the number of Adaptation Team members in attendance ranging from 8 to 11. Table 2Overview of adaptation team meetings and activities.DateDescription N Goals and Agenda05/06/2025Meeting 1Virtual10Introduction of meeting participantsBuild a shared understanding of current discharge process (part I)Literature review on evidence-based hospital discharge interventions and patients with NELP at hospital dischargeBrainstorm needs and strengths of patients with NELP relevant to hospital discharge05/13/2025Meeting 2Virtual11Review summary of last session and interim inputBuild a shared understanding of current discharge process (part II)Explanation of Core Functions and Form model (using every day example)Discuss list of core functions and brainstorm additional ones needed for populations with NELP05/20/2025Meeting 3Virtual8Review summary of last session and interim inputBrainstorm forms of a hospital discharge intervention for patients with NELP including: (1)Forms that correspond to a single core function (the “what”)(2)Cross-cutting forms that specify the means of delivery of other forms (the “who, when, where, and how”)*VotingAsynchronous9Rank importance of core functionsRate feasibility and impact of each core functionRate feasibility and impact of the crosscutting forms06/02/2025Meeting 4Virtual10Review results of group votingOpen discussion of proposed design of pilot intervention at Boston Medical CenterCFF, Core functions and forms model; NELP, Non-English language preference.*Initially planned on voting during Meeting 2 and 3 however based on feedback from Adaptation Team members this was made asynchronous after Meeting 3 to allow further time for deliberation. Overview of adaptation team meetings and activities. (1)Forms that correspond to a single core function (the “what”) Forms that correspond to a single core function (the “what”) (2)Cross-cutting forms that specify the means of delivery of other forms (the “who, when, where, and how”) Cross-cutting forms that specify the means of delivery of other forms (the “who, when, where, and how”) CFF, Core functions and forms model; NELP, Non-English language preference. Initially planned on voting during Meeting 2 and 3 however based on feedback from Adaptation Team members this was made asynchronous after Meeting 3 to allow further time for deliberation. Ranking of core functions In the ranking exercise, participants ranked “understanding and addressing priorities of the patients” as the most important core function (mean rank = 13.7, out of 15) followed by “assess and address linguistic and cultural needs” (mean rank = 11.8) and “involve caregiver” (mean rank = 10.9). Lower ranked core functions included “promoting self-care” (mean rank = 5.1), “ensure transmission of information to primary care provider” (mean rank = 4.3), and “promote inpatient continuity” (mean rank = 2.1). Rating core functions When Adaptation Team members rated the impact of each core function, “understand and address patient priorities” and “involve caregivers” received the highest impact ratings (mean impact score = 8.9 for both, out of 10) and were also rated as highly feasible (mean feasibility scores = 8.8 and 8.3, respectively). The next most impactful core function was “assess and address linguistic and cultural needs” (mean impact score = 8.4), but slightly less feasible (mean feasibility score = 7.6). Core functions rated as having low feasibility of achieving included “ensuring inpatient continuity” (mean feasibility score = 4.3), “empower patient to navigate the health system” (mean feasibility score = 5.7), and “promote self-care” (mean feasibility score = 5.8). These results are summarized in the bubble plot in Figure 3 (ranking in Supplementary Appendix). Quadrant plot of feasibility and impact ratings for core functions of hospital discharge interventions for patients with non-English language preference (NELP). Each bubble represents one core function, positioned by mean feasibility (x-axis) and mean impact (y-axis). Bubble size reflects the standard deviation, with larger bubbles indicating greater variance in ratings. Rating forms For cross-cutting options on who should deliver the intervention (Figure 4a), physician delivery was rated highest for perceived impact (mean impact = 9.5) but lowest for feasibility (mean feasibility = 4.8). Nurses, social workers, and community health workers were also seen as highly impactful (mean impact range 8.1–8.5) while being considerably more feasible (mean feasibility range 7.3–7.4), suggesting these non-physician roles offer a better balance of effect and practicality. Feasibility and impact ratings of cross-cutting forms of hospital discharge interventions for patients with non-English language preference (NELP): who delivers the intervention (a), when it is delivered (b), and how the language barrier is overcome (c). For decisions about timing and location of delivery (Figure 4b), Adaptation Team members rated in–person approaches as having greater impact than remote modalities, and models that included contact both during the hospital stay and after discharge as more impactful than those limited to only one time period. The highest impact model was in-person contact during admission plus post–discharge follow–up (mean impact = 9.4), but it scored low on feasibility (mean feasibility = 4.3). By contrast, a remote-only post–discharge- model (in which interventionists connect with patients via phone or video rather than in person) received the highest feasibility rating (mean feasibility = 8.3) but a lower impact score (mean impact = 6.6), illustrating a clear tradeoff between expected impact and practicality. For crosscutting options for achieving language concordance with diverse NELP participants (Figure 4c), team ratings favored in-person, language-concordant delivery as most impactful (mean impact = 9.4), closely followed by in-person delivery using a live interpreter (mean impact = 9.0). Phone delivery by a language concordant provider was rated as less impactful than in-person care (mean impact = 8.1) but more feasible (mean feasibility = 8.3). Remote delivery using an interpreter by phone or video scored lower on impact (mean impact = 6.8 and 7.5, respectively), indicating that language-concordant providers—especially in person—were seen as the most effective option, while telephone delivery by a concordant provider offered the best balance of impact and feasibility. Intervention forms Table 3 links the specific forms suggested by the Adaptation Team for four of the Core Functions. For understanding and addressing patient priorities, examples include structured tools (such as a goalsetting worksheet or a strengths/needs assessment) and standardized documentation of patients' top three priorities and linguistic/cultural preferences. To promote disease understanding, suggested forms include language-concordant written instructions, links to educational videos in preferred language, peer educators, and post-discharge phone calls to reinforce instructions. To involve caregivers, recommendations focused on documenting the primary caregiver and scheduling discharge teaching when the caregiver can attend. To facilitate post-discharge communication, proposals included providing a direct post-discharge phone line, creating an online question forum, enrolling patients in the patient portal during admission, and offering hospital lobby computers for messaging access. Table 3Example of specific intervention forms suggested by adaptation team linked to core functions.Core FunctionSpecific FormsUnderstand and address patient priorities- Create a document for patients to brainstorm health goals- Include in standardized documentation patient's top three priorities- Include in standardized documentation linguistic and cultural preferences- Develop a “strengths and needs” patient assessment toolPromote understanding of disease- Offer language concordant written discharge instructions- Provide links to educational videos in preferred language- Offer education from a peer- Telephone call post-discharge to reinforce written discharge instructionsInvolve caregiver- Document patient preferences around most important caregiver- Schedule discharge teaching at a time caregiver can be presentFacilitate post-discharge communication- Provide a number to call after discharge that gives direct access to a provider- Create an online post-discharge question forum- Actively enroll patients in MyChart while admitted- Have computers in hospital lobby for patients to use for post-discharge messaging in case they do not have access to technology Example of specific intervention forms suggested by adaptation team linked to core functions. Discussion The need for improved approaches to adapting complex multi-component interventions is well documented (86). In this manuscript we describe how the CFF model, integrated with the ADAPT methodology, guided our adaptation of hospital discharge interventions to better meet the needs of patients with NELP (11, 26). Framing adaptation around core functions (the causal processes that must be preserved to achieve intended effects) and adaptable forms (the context-sensitive ways to deliver those core functions) allowed us to translate shared functions across hospital discharge EBIs into delivery options tailored to those with NELP patients. Although interest in the CFF model is growing, prospective, stakeholder engaged applications remain-limited (87–90). Unlike prior applications that described adaptations retrospectively, we developed a stepwise, prospective process for engaging stakeholders to identify and prioritize core functions and select context appropriate-forms (91–94). In this way, the CFF model complements ADAPT: ADAPT provides the steps for assessing fit and documenting changes, while the CFF model supplies explicit guidance on what to preserve (functions) vs. what can be varied (forms such as activities, materials, and delivery modes) to maintain mechanisms of effect across contexts (11, 95). A novel feature of our methodology is applying the CFF model across a family of related EBIs rather than to a single program. We treated intervention adaptation and development as a continuum: infer core functions from existing hospital discharge interventions, prioritize those most likely to benefit the new target population, and then choose context-appropriate forms to deliver those functions. This function-to-form approach shows how shared mechanisms across different EBIs can be reused and tailored, enabling coordinated, scalable adaptations for interventions with common goals and components. After piloting the adapted intervention, we plan a hybrid effectiveness–implementation trial to confirm clinical benefit and continue evaluating implementation outcomes. Because our adaptation combines mechanisms and delivery features from multiple programs, such a trial is warranted in this case, but may not be in another application of the methodology where adaptations are less transformative. To enhance operational usefulness, we subdivided forms into specific forms and crosscutting forms. This clarifies “form” by separating what is delivered from who delivers it, when/where, and how. Specific forms are concrete components (e.g., written discharge summaries, post-discharge calls); crosscutting forms are implementation features that span components (e.g., deliverer, timing, communication mode, language modality). This distinction was useful where a single core function could be achieved via multiple component-delivery combinations. Making specific and crosscutting forms explicit highlighted implementation decisions often omitted from intervention descriptions and prompted systematic consideration of linguistically relevant adaptations (e.g., in-person vs. remote delivery and how to achieve language concordance). These choices often determine-whether an intervention is accessible and effective for NELP patients. A key drawback of the CFF model is that it does not systematically prompt adapters to anticipate unintended consequences of changing forms. Frameworks such as Model for Implementation Design and Impact (MADI) emphasize mapping downstream “ripple” effects and documenting potential mediators and moderators when delivery features are altered (21). For example, replacing in person, language-concordant delivery with interpreter-mediated-phone encounters may still convey key discharge information (core function), but can also reduce patient engagement, erode trust, and limit nonverbal communication (96). Those downstream effects may blunt the intervention's mechanism (e.g., patient activation or comprehension) and ultimately reduce its effectiveness counter to the intention of adaptation efforts. When correctly applied the CFF model should help adapters know through measurement of core functions if it was achieved or not; without complementary attention to causal pathways it may not explain why a form change produced that failure. A central challenge for the CFF model and other mechanism focused adaptation frameworks is reliably identifying the mechanisms themselves. Guidance calls for explicit attention to theory or mechanisms of action and for preserving “core components,” yet many trials do not define or justify those elements (1). In our case study we relied on a preliminary list of core functions derived from a structured literature review and additional input from the Adaptation Team, similar to others (97). An alternative and more rigorous approach would be to use a formal consensus process (e.g., Delphi) to engage a broad group of experts in hospital discharge intervention including social workers, physicians, and nurses to define functions (93, 98). As the field advances, aligning terms such as “core components,” “mechanisms,” and “core functions” across intervention development, implementation science, and adaptation will improve consistency and guide methodological choices (88, 99, 100). An additional limitation concerns the level of abstraction at which core functions should be specified, as prior applications of the CFF model have varied in granularity. For example, in applying it to the patient-centered medical home, Perez Jolles et al. nested multiple core functions within an overarching principle (e.g., “accessible care”) (26). In our application, some elements, such as “involve caregivers,” could reasonably be conceptualized as higher-level principles rather than core functions, with more specific operationalizations articulated beneath them. Alternative structuring choices may have resulted in a different set or hierarchy of core functions. Future work should further clarify guidance on how to define and operationalize core functions to enhance consistency and comparability across studies. Two additional limitations should be noted. First, we did not employ a formal Delphi process with predefined consensus thresholds. This was a deliberate choice: the Delphi technique is expressly designed to prevent direct interaction among panelists, whereas dialogue among patients, caregivers, and clinicians was central to our codesign approach. Ranking and rating exercises were therefore used to surface divergence and focus discussion rather than serve as decision rules. We acknowledge this limits comparability with consensus derived through established Delphi methods (101). Future efforts could consider the nominal group technique or equity-focused modifications to the Delphi process to promote equitable participation (102–104). Second, the Adaptation Team skewed female and predominantly comprised health care workers—including several members recruited to represent patient and caregiver perspectives—narrowing the range of lived experience captured and potentially shaping which core functions were prioritized and how they were interpreted. This was partially mitigated by the team's composition: 54% spoke a language other than English at home and 85% had caregiving experience. Nonetheless, professionally embedded perspectives may not fully reflect those of patients with NELP navigating discharge without clinical knowledge or institutional familiarity. The approach we present here has important implications for health equity. Patients with NELP face many barriers to EBIs and have disproportionately high risk of communication failures at discharge and adverse outcomes such as unplanned readmissions (32, 48), yet have largely been excluded from prior discharge trials (40, 41). Creating entirely new interventions for each language group would be costly and underuse existing evidence (105). Health disparities scholars have warned of “sideways progress,” where repeated, intensive adaptations for each demographic subgroup consume resources without producing broadly scalable solutions (106). Our approach seeks efficiency and real-world-relevance by (1) building on proven discharge EBIs and making targeted adaptations needed for NELP populations and (2) adapting for the broader NELP group rather than designing separate programs for each language or cultural subgroup, aligning adaptations with diverse clinical settings to improve feasibility and scalability (10). Conclusions Integrating the CFF model with the ADAPT process offers a practical and theory-grounded approach for equity-focused adaptation of complex, multicomponent interventions and effectively facilitated adaptation by drawing from the numerous hospital discharge interventions tested to date. By specifying measurable core functions and distinguishing between specific and cross-cutting forms, this combined approach helps implementers preserve intervention mechanisms while tailoring delivery to the linguistic and cultural needs of populations with NELP. Applying these methods to a family of related discharge interventions, rather than to a single program, illustrates how shared mechanisms can be translated into context-sensitive delivery options that are both actionable for health systems and amenable to rigorous evaluation. For practice, the core functions–forms distinction gives frontline teams a transferable way to adapt the surface features of a discharge intervention to the patients in front of them while safeguarding the elements that drive outcomes, rather than relying on ad hoc or improvised adjustments. For health equity, this approach makes rigorous adaptation feasible at scale for populations historically excluded from discharge trials, helping ensure that the benefits of proven EBIs reach NELP patients rather than stopping at the boundary of the languages and settings in which they were first tested. Future work should refine guidance on defining core functions and test this approach across diverse settings and populations.