Rigorous projects and grants integrating human-centered design and implementation science: 8 tips for success

Abstract

Background

Researchers and funders increasingly see potential for integrating human-centered design (HCD) and implementation science (IS) approaches to enhance the design and uptake of health interventions and implementation strategies. These approaches offer complementary strengths. HCD prioritizes developing compelling, intuitive, and adoptable innovations grounded in user and contextual needs; IS focuses on ensuring evidence-based practices reach and work for their intended populations. Integrating HCD and IS creates challenges that existing IS proposal writing guidelines do not address. HCD’s inherent uncertainty–since final designs are developed through iteration rather than pre-determined–can unsettle researchers and funders accustomed to more defined endpoints. Variation in how HCD is understood adds further difficulty when collaborating across disciplines or communicating with funders. There are currently no published recommendations to help researchers design strong studies integrating HCD and IS and communicate them effectively to funders.

Discussion

We present tips to support both rigorous study design and effective proposal writing that integrate HCD and IS. This draws from our experiences leading, supporting, and reviewing both funded and unfunded projects. Our considerations are informed primarily by experiences with teams in the United States in the mental health and behavioral health domains; however, we have found our findings also apply to projects globally and in other health sectors. Our intent is that this guide will advance good research that leverages both approaches and is viable for funding. We also hope that the tips will support funders in developing solicitations and evaluating proposals.

Conclusion

There is untapped potential for innovative health research integrating HCD and IS, but realizing it requires both rigorous study design and an ability to attract funding. We offer practice guidance to help researchers and funders develop quality projects and proposals, and we encourage others to build on these recommendations.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s43058-026-01035-5.

Contributions to the literature.

• Human-centered design (HCD) offers tools for systematically collecting input from affected individuals and communities and emphasizes iterative improvements to innovations based on user input. There is increasing interest in integrating HCD and IS to advance implementation of evidence-based practices.

• HCD’s defining features of flexibility and iteration based on user needs are unique strengths but can be challenging to convey to team members and funders of traditional implementation research. Varying familiarity and interpretation of HCD can create an additional hurdle when collaborating with interdisciplinary team members and communicating with funders.

• We outline tips to strengthen research that integrates HCD and IS and help teams secure funding to support this work.

Introduction

Over the past two decades, there has been an expansion in the use of HCD across health topics and to support the design of digital (e.g., mobile health apps) and non-digital (e.g., behavior change interventions) solutions [1]. Human-centered design (HCD)—an approach to creating compelling, intuitive, adoptable, and engaging products, systems, and services based on the needs of individuals and systems [2]—has been increasingly applied to create person-centered health services [3]. There are varying definitions and models of HCD, which reflect its origins and evolution. Design traditions that flow into and through HCD include human factors engineering, human-computer interaction (HCI), user-centered design, and design thinking [1]. An underlying premise across these design traditions is that design should be driven by human experiences and that solutions are enhanced through iteration based on user feedback. In this paper, we use HCD as an inclusive term and flexible approach to leverage the complementary methodological and measurement approaches of different design traditions and to reflect the diverse applications of design in health research [1].

Justification for integrating human-centered design (HCD) and implementation science (IS)

HCD and implementation science (IS) are complementary, with literature articulating the links between the two approaches and utility of HCD to identify usability issues with existing interventions [4], adapt and redesign evidence-based programs, practices, and policies (EBPs) and implementation strategies [5, 6], and assess implementation determinants and outcomes [7]. While significant advances have been made to implement EBPs and study their implementation in practice, opportunities remain to more consistently address user needs, improve implementation outcomes (e.g., beyond acceptability to penetration and sustainability), and expand health impact [3, 8]. HCD can be leveraged to systematically identify user and context needs, ideate and prototype innovation solutions based on needs while preserving effective components (i.e., fidelity-consistent adaptation) [9], and evaluate innovation-user and innovation-contextual fit [10]. HCD can be applied to both the design or adaptation of an innovation to fit a relevant context, and to designing how it is delivered in practice (e.g., its service delivery model and implementation plan) [11]. Applying HCD during implementation can surface problems and inform solutions more quickly than waiting for a trial to end, resulting in greater efficiency and impact for individuals, systems, and future IS research [12, 13]. Importantly, HCD offers tools to facilitate engagement of users throughout the design process so that users are actively involved in tackling implementation challenges.

HCD and IS funding landscape

Even well-designed projects require convincing funders that they are worthy of investment, and teams must navigate an underdeveloped funding landscape for research that integrates HCD and IS. National Institutes of Health (NIH), US Veterans Affairs, and private foundations have set expectations for those seeking funding for IS research [14–17]. Proctor et al.’s “Writing implementation research grant proposals: ten key ingredients” is a formative resource for those seeking to write competitive IS proposals [18]. These recommendations have been used to evaluate proposals [19] and for grant writing trainings [20]. Another resource, Brownson et al.’s “Concocting that Magic Elixir: Successful Grant Application Writing in Dissemination and Implementation (D&I) Research”, identifies 12 core competencies that should be covered in NIH D&I proposals through a card sorting exercise with IS researchers [21]. In contrast, the HCD funding landscape is less established, a likely reflection of its interdisciplinary origins and practice. It is rare to see funding solicitations that explicitly request an HCD approach, which contributes to a lack of understanding of funders’ expectations. Currently, there are no publicly available analyses of HCD proposals or guidelines. Funding calls and grants that involve some form of community engagement in health research are widespread, highlighting the relevance of HCD to researchers [22–24]. However, expectations and measurement of community engagement vary across funders [23, 25, 26]. Funders can be a barrier to meaningful community engaged work [27]. An analysis of NIH-funded IS grants in 2020 and 2021 revealed that 87% of grants included such form of community engagement, however, the majority of grants involved community consultation not partnership [23]. As result, there is an opportunity to strengthen support to funders and researchers seeking to do HCD and IS health research.

Challenges to integrating HCD and IS

HCD and IS share conceptual convergence in aims but can diverge in disciplinary framing. Table 1 compares HCD and IS and summarizes challenges [28, 29]. First, despite many shared values and priorities, HCD and IS have different foundational orientations. HCD prioritizes innovation grounded in user needs and context, and IS emphasizes what’s needed, what works, and improving fit to context. IS traditionally assesses fit with users and context at structured timepoints, although there are frameworks that support more flexible adaption of EBPs including the Dynamic Adaptation Process [30], Learning Health Systems [31], and Dynamic Sustainability Framework [32]. These IS approaches, which emphasize more responsive improvement to EBPs, are aligned with HCD’s iterative design process. HCD’s focus on innovation may initially overshadow focus on IS outcomes. HCD’s experimental problem solving can take precedent over translating solutions into real-world use, a strength of IS [28]. However, findings from HCD may accelerate attaining implementation outcomes. Both fields emphasize different but related concepts to measure impact. HCD tends to dedicate more time to Discover/formative research and iterative design, and testing at a smaller, quicker scale. Finally, HCD historically focuses on designing for and with closely affected people, whereas IS considers users across multiple levels that enable implementation [33].

Table 1.

HCD and IS have different orientations that affect integrating the two approaches for proposals

	HCD	IS
Top priorities	User needs, innovation	Implementation outcomes (acceptability, adoption, appropriateness, costs, feasibility, fidelity, penetration, sustainability), starting with what works
Evidence/Values	Engagement over time and usability, with adaptations/iteration, knowledge specific to design solutions	Coverage and quality of services, implementation outcomes, health outcomes, generalizable knowledge
Timeline	Can take a long time depending on the nature of user engagement during formative research phases, types of design solutions, and needed iteration.	Can be quicker since “thing” is already developed and there is interest to move quickly for deployment. Adaptation and iteration can happen after the “thing” is tested.
Who	End users	Multi-actors and systems

To leverage HCD and IS’s complementary strengths and different orientations, this article outlines tips for developing rigorous research and proposals that integrate both approaches for health research. Our recommendations are influenced by experiences leading and supporting such projects to improve evidence-based mental health treatments through the NIH-funded University of Washington Advanced Laboratories for Accelerating the Reach and Impact of Treatments for Youth and Adults with Mental Illness Center (UWAC). UWAC runs a bootcamp focused on writing NIH and other U.S. federally-sponsored proposals integrating HCD and IS. We also draw on our experiences writing and reviewing proposals for funders, including NIH, National Science Foundation, US Agency for Healthcare Research & Quality, and Robert Wood Johnson Foundation. One author (TA) also leverages experiences working on projects integrating HCD and IS for nutrition in East Africa. Both funded and unfunded projects have informed these tips. While we present tips sequentially, tips are interdependent and interconnected.

Tip 1: What roles do HCD and IS have in your project?

Teams should determine whether HCD is appropriate and, if so, the role of HCD and IS in the project. HCD can be unsuitable depending on project goals, context, culture, and funding mechanism. HCD may also be challenging to use for projects where solutions must fit within systems that prioritize vertical decision-making rather than participatory processes. A project team may set themselves up for future difficulties if they push an HCD approach with individuals who do not support HCD priorities. If HCD is appropriate, teams should establish consensus on project goals and the role of HCD. Is it an IS research project that uses HCD? Is it a design research project that is informed by IS? Is it a blend? Identifying this will help build the case for integrating HCD and IS.

IS research projects informed by HCD: HCD is a tool for redesign and formative research

IS focuses on identifying what is needed and what works, then ensuring that it can work reliably at scale and over time for different populations. The primary active ingredients of the “thing” (e.g., interventions, practices, innovations) [34] are typically supported by existing evidence from efficacy and effectiveness research [35] which might be further developed via effectiveness-implementation hybrid studies [36]. Effectiveness-implementation hybrid studies are only recommended in situations where there is at least indirect evidence for the EBP (e.g., effectiveness data from a different but related context) [36]. IS projects can use HCD to support redesigning interventions and implementation strategies, ensuring that both are acceptable, appropriate, and feasible. For example, UWAC’s three pilot (NIH R34s) and large (NIH R01) studies—each redesigning EBPs for community mental health settings—are IS projects that use HCD to redesign interventions and implementation strategies [5]. These projects are using effectiveness-implementation hybrid studies to test redesigned versions against their respective unadapted versions. Projects should prioritize design-related evidence gaps for new data collection. HCD can also be an indispensable formative research tool to better understand implementation determinants, opportunities to optimize implementation strategies, evaluate implementation and HCD outcomes (including usability), and explore future design solutions that may be currently infeasible [7, 37]. For example, HCD supports formative early-stage research to develop an intervention or assess participants’ perceived usability of an intervention.

HCD research projects informed by IS: IS to promote sustainment and scale across systems

HCD focuses on creating knowledge that can help create better solutions and inform future design work. There are several design research interpretations including research for design and research through design. Research for design describes research to generate evidence that informs a design, such as user research [38, 39]. Research through design treats the design process as a mode of inquiry [38, 40]. An example of this in health service research could be redesigning an intervention for a specific community and identifying considerations when adapting the intervention in the future for other communities. HCD projects should have evidence suggesting that there is a problem that design can fix and propose appropriate methods. Because HCD research can struggle with translating innovations into real-world practice, IS can be integrated to consider implementation contexts for maximal impact. HCD research on designs that may be possible in the future with changes in technology, policy, or infrastructure can also benefit from using IS approaches to unpack system level changes to enable innovations.

Tip 2: What are your design problems and questions?

Projects should have strong research questions and complementary design questions (Table 2). Defining design problems and design questions is a crucial step to scoping a project’s design work, selecting appropriate methods, managing project uncertainty, and assembling a project team with the right expertise. Design problems state what problem your design is intended to solve for users [41]. Strong design problems identify direct and indirect users, problems they face, and why the problem is important. Direct users directly interact with an intervention and indirect users are people affected by an intervention [5]. Design problems should represent a balance of scope; avoid being too specific so that there is space for iteration based on user research and avoid temptations of taking on too many problems. For example, a design problem of poor medication adherence for elderly patients due to pill bottle font size is very specific and points to a specific solution – changing the font size. However, poor adherence could be due to other factors like cognitive decline or physical limitations opening a pill bottle. The complexity of health services and related structural inequalities can draw teams toward pursuing a broad problem space. For resource-limited projects, teams should focus on mitigating negative consequences of inequities. Design justice recommends considering asymmetric distribution of benefits and harms when framing problems [42]. A design problem can be described in a proposal’s background section and can be based on completed preliminary research or gaps in the field.

Table 2.

Distinctions among design problems, research problems, design questions, and research questions

	Design Problem	Design Question	Research Question
Description	What problem a design is intended to solve for a group of people	A focused query on what a design is intended to address for a group of people	A focused query to fill a knowledge gap
Example: Psychosocial interventions	Quality of evidence-based psychosocial interventions can vary based on provider access to supervision and guidance	What are tools that can be designed to support providers in delivering quality evidence-based psychosocial interventions?	What is the effectiveness of different types of supervision and guidance available to help clinicians deliver quality evidence-based psychosocial interventions?
Example: Integrated management of acute malnutrition (IMAM)	IMAM is a global evidence-based program to prevent and treat wasting among children, and implementation has not reached its potential due to insufficient adaptation to community needs in Tanzania	How might communities, health providers, and policy makers collaborate to redesign IMAM to improve program coverage in Tanzania?	What are factors influencing implementation of IMAM by different actors in Tanzania?
Example: Medication adherence	Older adults can take their medications inconsistently and incorrectly	How might we support older adults in taking their medications as instructed?	What strategies do older adults currently use to manage their medications, and where do these strategies break down?

Design questions state what your design is trying to solve in the form of a question. One common approach is to use a “how might we” format to develop design questions [43]. If using this format, it is important for teams to reflect on who is being considered the “we”. If the “we” is only the research team, this can reinforce power dynamics of designing for rather than with. Design justice emphasizes how affected communities should be included and control design processes that are intended to benefit them. This also recognizes that those affected have frequently already developed solutions or workarounds [42]. Design questions may evolve over time in response to research. In HCD, research questions are used to identify existing knowledge or generate knowledge that can help answer design questions. The answers to design questions are solutions or approaches, whereas the answers to research questions are understanding or new knowledge.

Tip 3: What methods align with your questions and context?

Integrating HCD and IS is an exciting opportunity to draw upon different quantitative and qualitative methods. In IS, there are numerous quantitative and qualitative methods used to understand intervention and implementation strategies and outcomes [44, 45], with prioritization of method feasibility and relevance to users [18]. Approaches used in HCD can be combined with IS methods to create a more holistic understanding of a user’s experience with a design. For example, rapid ethnographic assessment, used in IS [46], could be combined with usability testing from HCD to observe challenges users experience when attempting to complete steps of an intervention. Some methods commonly used in HCD are more novel to IS and health research (e.g., prototyping, journey mapping), and other methods (e.g., interviews, focus groups) are more familiar yet indispensable to HCD [29, 47–50]. Given the volume of methods that can be used in HCD and limited existing guidance on how to select and adapt methods, it can be difficult deciding what methods to propose for a project. The most important criterion for method selection is alignment: methods should match research and design questions, fit the user context, and be feasible for the team.

Methods used in HCD

We recommend that teams review existing design method compilations [50–52], then investigate promising methods in the literature for use cases and adaptations. A single research or design question may be answered by multiple methods, and a single method could answer multiple questions. All methods can be subject to observational or social desirability bias, highlighting the importance of employing multiple types of methods to understand a user’s relationship to a design. Table 3 presents a template to help organize methods for a project and facilitate team member discussion.

Table 3.

Method selection template

Design/User Research Question	Method(s)	Data to be Collected	Analysis Plan (and How it Relates to the Design Question)
-	-	-	-

Methods commonly used in HCD can be categorized based on the type of data they generate (behavioral/attitudinal, quantitative/qualitative) and design phases (exploratory/generative/evaluative) [51, 53]. Behavioral (what people do) and attitudinal (what people say) data can be paired to understand a user’s beliefs and actions with a design [53, 54]. There can be a disconnect between people’s perceived and observed behavior [55, 56]. For example, usability testing involves observing people using a design (behavioral) and often includes interview questions to understand the “why” behind a person’s behavior (attitudinal) [57]. Interviews and focus groups can be used to ask people about their opinions (attitudinal) or how they interact with a design (more behavioral). Identical methods can be used at different stages of the design process, but with varying objectives. For example, interviews can be used during an exploratory phase cognitive walkthrough [58–60] to understand barriers and facilitators to an existing intervention, as part of a generative phase co-design session to iteratively refine designs [51, 61], or as part of an evaluative phase usability evaluation to assess how well users can achieve their goal with a design [51, 57]. While a comprehensive discussion of different methods is beyond the scope of this paper, we include recommendations on prototyping, testing, and participant sampling in Supplemental Files 1–2.

Common mistakes

Common mistakes that we see lose sight of this principle: proposing too many methods suggests insufficient attention to feasibility, while defaulting to a narrow set of familiar methods misses opportunities to better address the questions at hand. Teams proposing a narrow set of methods based on what they are familiar with is a missed opportunity to embrace a broad range of methods. For example, co-design may be more aspirational than feasible if there is limited interest in sharing decision-making [61]. Misalignment can also stem from improper application of methods. For example, creating personas that are not based on data from actual intended users can lead to designing for an imaginary user constructed from assumptions and stereotypes [62–65]. Teams should assess if the methods chosen fully cover research and design questions with minimal overlap [1]. If teams do not have preliminary evidence that methods proposed are acceptable and appropriate with participants, teams can include several options or contingencies if methods do not work out. Although elaborating on the feasibility of methods can be difficult with proposal space limitations, teams could frame proposed design methods as preliminary and subject to change with additional consultation of communities and cite examples of similar design methods used in other projects.

Tip 4: How will you plan for uncertainty and communicate your plan?

While IS has long recognized uncertainty in how EBPs are delivered over time—reflected in concepts like “program drift” and “voltage drop” [32], unplanned and reactive adaptations in the Framework for Reporting Adaptations and Modifications – Expanded (FRAME) [66], and dynamic frameworks for EBP adaptation—HCD introduces an additional and distinct source of uncertainty: the design itself is unknown at the project’s outset. This kind of uncertainty—not knowing the solution, only the process for arriving at it—can be unnerving to teams and funders accustomed to defined endpoints and concrete solutions. Therefore, it is important to consider the role of uncertainty in any project that uses HCD to alleviate perceived risks, as this can affect all aspects of study design and securing funding. Uncertainty and risk proposed must match collaborators’ and funding mechanisms’ risk tolerance and team resources.

Avoid addressing uncertainty with predetermined solutions

One mistake made is projects can claim to use HCD but focus on a predetermined design solution with limited willingness to explore different design solutions. While focusing on a predetermined design solution may seem more concrete to team members and funders, this is against the intention of using HCD to design based on user needs. This is most problematic when HCD serves only to legitimize predetermined solutions through performative user research. Projects using HCD should avoid promising specific design solutions until user needs and constraints are understood (e.g., only proposing designing an app without evidence that people want this). If teams intend to focus on predetermined solutions without evidence that these are desired solutions, it is important to acknowledge that the project is not truly user-centered from the start. However, HCD can still be used to understand intervention usability and engagement. Projects involving predetermined solutions without an evidence base may be more suitable for effectiveness-implementation hybrid designs to assess clinical effectiveness and implementation [36].

Aligning funder expectations with HCD

Teams should engage funders early to assess alignment with funding priorities and comfort with HCD’s iterative nature. Funders who are only interested in seeing specific predetermined solutions may be less keen on funding projects with HCD. For example, funder feedback on one unawarded grant included disappointment that we had not proposed their preferred solutions. This prompted reflection on how we could have better communicated potential design solutions based on emerging user needs. UWAC outlines research expectations for our pilot project (NIH R03s) request for proposals (Supplemental File 3) [67], which is intended for investigators new to integrating HCD and IS. Common challenges among the proposals we review suggest potential discomfort with uncertainty, including focusing on predetermined solutions, and issues with project scope.

Frame uncertain design directions in the context of evidence and potential outcomes

Drawing on guidance for communicating transformative [68] and risky research [69]—which recommend grounding uncertain work in what is known, articulating decision points, and outlining potential outcomes — we recommend that teams address uncertainty by building projects around a design process with embedded decision points triggered by emerging insights. From the start, projects should outline what is known—including related evidence-based “things”, implementation challenges, and user characteristics—and what’s unknown. Core questions are: (1) What’s the health issue?; (2) What are the existing evidence-based “things”?; (3) Why are “things” suboptimal(ly) implemented?; (4) What types of solutions may emerge to address implementation challenges? Answering these questions can help you develop a list of key decisions the team will need to make (and when), potential outcomes, and how you will proceed based on outcomes. Table 4 includes excerpts from the UWAC Trauma-Focused Cognitive Behavioral Therapy (TF-CBT) [5] proposal’s Significance section, which addresses the questions above. The proposal outlines potential redesign solutions based on barriers identified in existing research, while not committing to any of these redesign solutions as grant deliverables. During co-design sessions, participants independently proposed developing a skills repository, guidance on alternative trauma narrative formats and scheduling around school schedules, a school readiness checklist, and a caregiver involvement decision tree. Many of these final solutions are thematically aligned with what the team included as potential solutions in the original proposal, while others are unexpected.

Table 4.

Addressing uncertainty in TF-CBT NIH R34 proposal

1. What’s the health issue?	More than two-thirds of youth in the United States experience trauma [70] and 1 in 5 will develop post-traumatic stress symptoms that impair their functioning [71].
2. What are the existing evidence-based “things”?	Trauma-focused cognitive behavior therapy (TF-CBT) has the strongest evidence of any clinical intervention for youth trauma [72, 73], but is rarely adopted in the most common setting for youth mental healthcare: schools.
3. Why are “things” suboptimal(ly) implemented?	Use of TF-CBT in schools is limited by (1) problems with its usability (e.g., rigid structures, complicated patient identification workflows) and (2) practitioner perceptions that some core elements (e.g., exposure) are not contextually appropriate for schools; these aspects hinder practitioner and student engagement with TF-CBT [74, 75].
4. What types of solutions may emerge to address implementation challenges?	Session length – shortening sessions to 20–30 min Caregiver involvement – flexible caregiver session scheduling and alternative caregivers Referral pathways – revised family engagement and referral pathways School specific consultation – certified TF-CBT trainer experienced working in schools and/or with school clinicians

Tip 5: What type of expertise do you need for a strong team?

Projects should demonstrate the capacity of a proposed team to carry out a project and achieve desired outcomes. The decision of who can best provide HCD expertise and support project needs should largely be driven by considering the role that design is expected to play in the project, which we emphasize in Tip 1. Table 5 outlines sample expertise that may need to be prioritized for a HCD and IS project. For HCD and IS grants submitted to traditional IS health funders, teams should have individuals with sufficient skills in IS and the health topic. For NIH, these skills are expected to be reflected in NIH biosketches. Teams and individuals trained in design may be skilled at different methods, and some may be better suited to a project than others. For example, a project that has design needs for a digital or computational solution would benefit from an individual with HCI or user experience (UX) design that is familiar with prototyping digital solutions. We encourage you to consult with team science literature for strategies on collaboration, capacity building, communication, and project management to integrate skills and expertise across team members [76–78].

Table 5.

Diverse and complementary skills are needed to create a strong team

Skill	Description
Health	• In-depth knowledge of health topic and active in research community • Clinical experience (if applicable) • Research skills (study design, data collection, analysis) • Understanding relevant ethical considerations
Implementation science	• Familiarity with existing implementation strategies and theories/frameworks for health innovations • Familiarity with implementation outcomes and their assessment and role in attaining subsequent more distal outcomes • Research skills (hybrid designs, data collection, analysis) • Understanding relevant ethical considerations
User experience research	• Proficiency in HCD user research methods and principles • Research skills (study design, data collection, analysis)
User experience design	• Proficiency in design methods and principles • Prototyping skills • Experience in expected design medium (application, print, service, video, etc.)
Systems science	• Ability to analyze and understand complex systems and their interdependencies • Consideration of user, technological, and community contexts in design decision-making • Proficiency in identifying and addressing system-level challenges and opportunities
Communication and facilitation	• Manage project • Document project decisions and findings • Liaise with key users and interested parties
Innovation and continuous improvement	• Openness to explore new research methods and design approaches • Drive culture of continuous improvement within the team

Table 6 summarizes strengths and risks of different options for design support. Teams that anticipate substantial design work might consider hiring a design firm, independent consultants, or full-time user researchers or designers. This can be an effective way to accomplish focused, high-quality work, but can be costly and include unnecessary non-negotiable additional services (e.g., project management). It is important to carefully write and review contracts with design firms or independent consultants to ensure that rates are within budget and to reach agreement on intellectual property. When design work is small and well-scoped and hiring a design firm or full-time staff is out of reach, undergraduate or professional master’s students may be a good fit. Undergraduate or master’s students studying UX can provide project support through a class project or an independent study. In these arrangements, students’ learning goals must come before the project goals, which adds some uncertainty: students may produce something that exceeds expectations, but they may also not deliver something that meets project goals. This arrangement works best for supplementary work rather than critical deliverables: student timelines, learning goals, and skill levels may not align with what a project requires at a given moment. Doctoral students can be valuable contributors to projects as emerging researchers developing their own expertise, especially when the work is aligned with the student’s research interests and training goals. In most HCD programs, faculty work closely with doctoral students in an apprentice model. For teams that include students, feedback and mentorship mechanisms should be established; allocating level of effort for a faculty member with HCD expertise can support this.

Table 6.

Filling gaps in design expertise requires weighing strengths and risks of different options

Group	Their Needs	Strengths	Risks
Design firm or consultants	• Remuneration	• Potentially high level of expertise • Contractually obligated to deliver end result	• Ownership of intellectual property (address through careful contracting) • May not be familiar with prioritizing certain features to support research goals (rather than overall user experience), which may not be aligned with project needs • Expensive
Design staff as part of the project	• Remuneration • Include work in portfolio	• Potentially high level of expertise	• Only large projects can support full time staff • Plan to support staff after project ends • Many universities cannot pay market rate
Design faculty (potentially with doctoral research assistant)	• Full-Time Equivalent (FTE) • Support for doctoral students (stipend, tuition, benefits) • Alignment with their research goals and vision	• Potentially high level of expertise • Familiar with some trade offs required in research prototypes, but may be unfamiliar with scale needed for successful projects in health sciences • Can support student training	• Attention can be fragmented (address with recruiting faculty supported by research assistant)
Hourly student or research graduate student	• Remuneration • Include work in portfolio	• Lower cost • Paid hourly • Compared to student projects, there exists a clear employment relationship with expected deliverables	• May need more mentorship • Student schedule may not align with project needs
Course project, for-credit research, or independent study	• Mentorship • Include work in portfolio	• Supports student training • Lowest cost	• Student educational goals come before project goals • May need more mentorship • At the end of the quarter/semester, they are typically done - even if the project is not (sometimes opportunity to then transition to hourly pay)

When the proposed project includes research questions that are of interest to design researchers, it may make sense to include them as co-investigators. For example, design researchers could be interested in research through design [39, 79], like developing new design methods or adapting existing design methods for a different context or population, or how design choices are tied to user behavior [80]. Design researchers, however, are unlikely to engage in projects to do the design work unless there are research questions of interest to them. This is akin to clinical researchers who are not typically investigators on projects to do clinical work. Nonetheless, during the project, design researchers may do some design work and clinical researchers may have some clinical encounters. Design researchers should be involved early in the problem definition process as they will bring their own research questions agenda to the topic.

Tip 6: What is your plan for engaging with users?

HCD and IS focus on users in complementary ways. Major IS frameworks—Consolidated Framework for Implementation Research (CFIR) [81], Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) [82], and Practical, Robust Implementation and Sustainability Model (PRISM) [83]—emphasize how multilevel individuals and systems influence health services. HCD offers techniques to incorporate perspectives and engage communities in different ways to design solutions, which can be incorporated in IS’ multilevel considerations. Research integrating HCD and IS should have a plan for identifying actors, the extent to which the team will engage with them, and how the team plans to incorporate feedback to answer design questions. One way to approach this is by conducting a stakeholder analysis, which assesses stakeholders’ behaviors, position, and influence on adoption of a project [84] and is recommended by Proctor et al. for IS proposals [18]. When conducting a stakeholder analysis, teams should consider actors’ relationship to a design solution, which may influence how actors are involved. Projects may tailor their methods to different groups to explore various dimensions of a design problem. Teams should also consider power dynamics among groups. For example, we have been involved in projects where co-design sessions are done separately with providers and patients, both to capture distinct perspectives each group offers and to give each the space to freely share experiences and perspectives.

Community-engagement continuum

Community engagement in health research exists on a spectrum from community informed (e.g. consultative) to community driven/led research (e.g. power lies in communities) [85]. HCD research that involves communities can be similarly characterized by this spectrum [86, 87]. A key decision in HCD and IS projects is the degree of community involvement and power-sharing. This choice connects with other tips in this article, particularly method selection. A mistake we have seen in projects is an inflation of the nature of community engagement. We encourage early discussions with collaborating communities, organizations, and the systems where designs will be implemented about their experience with participatory approaches and comfort with collaborative decision-making. Teams should also be prepared to pivot the nature of community engagement to protect participants’ safety and confidentiality. For example, Fernández-Sánchez et al. describe strategies for protecting participant safety in community-engaged research with migrants in the U.S. during times of political uncertainty [88], which are transferrable to other contexts.

Tip 7: How will you measure success?

Measurement plans should outline constructs to be measured, conceptual models, and an analysis plan. IS measures characterize implementation success by considering effectiveness and contextual factors (Table 1) [89, 90]. HCD measures emphasize a person’s experience, emotions, and attitudes towards a design, and the alignment of a design with a person’s goals [91]. Depending on project scope and size, some projects may also propose measuring clinical impact. The aim of HCD and IS measures is similar—evaluate and improve designs [90, 91]—and can be measured quantitatively and qualitatively. The UWAC theory of change argues that successful adoption of EBPs is limited due to usability, engagement, and appropriateness [5]. Usability and engagement draw most from HCD, while appropriateness draws from IS. These measures are the direct target of UWAC affiliated projects when redesigning EBPs. Operationalizing HCD and IS measures alongside each other in our projects has surfaced practical considerations, including how to sequence measurement across design phases and how to select instruments sensitive enough to detect changes from iterative redesign [5]. This is merely a sample of potential HCD and IS measures a team can choose from. For IS measures, we recommend teams consult with the substantial existing literature on IS outcomes, measurement approaches, and psychometric properties to determine relevant IS measures. Table 7 lists a sample of HCD outcomes, utility, and measurement approaches, which we detail further in a previous paper [1].

Table 7.

Sample HCD outcomes, motivation, and measurement approaches used by UWAC projects

Outcome	Definition	Examples of utility	Examples of quantitative measurement	Examples of qualitative measurement
Usability	Extent to which specific people can achieve specified goals of effectiveness, efficiency, and satisfaction with a service or system [92]	Explore usability issues of unadapted interventions and implementation strategies. Identify avoidable and costly usability issues that can be addressed before larger-scale implementation of interventions and implementation strategies.	System Usability Scale [93]/Intervention Usability Scale [94]/Implementation Strategy Usability Scale [59] scores, task completion rates during usability testing	Heuristic evaluation [95], insights from observed task completion and interviews during usability testing [57]
Engagement	Degree of user participation and enthusiasm for the aspects of clinical interventions and implementation strategies that require user involvement [96]	Explore factors that lead individuals to drop out of an unadapted interventions and implementation strategies. Identify whether design changes improve adherence to interventions or implementation strategies.	User Engagement Scale [97], UWAC Engagement Measure which is modelled after the Patient Responsiveness Scale [98]	Insights from observations and interviews [96]

Tip 8: What safeguards and accountability mechanisms will you include to catch unintended harms as your project unfolds?

Teams have an ethical responsibility to ensure that projects embody a responsible use of design, with equitable distribution of benefits and harms that do not disproportionately burden vulnerable groups. While those that use design come with positive intentions, there are numerous examples of design processes exacerbating inequities [42, 61, 99–101]. We developed self-assessment questions for teams to encourage appropriate and effective use of HCD [1]. Examples of potential safeguards and accountability activities in HCD research include periodic evaluation points during a project life cycle to confirm a design is still appropriate and safe, and protocols for participants and team members to report and escalate a potential unintended harm with a design.

Positionality and reflexivity

In recent years, there has been increased attention towards how structural racism and power imbalances influence implementation and health equity [102–106]. There are numerous calls to address these issues by design researchers [107–111]. Positionality encompasses a researcher’s social and cultural identity and values, which can influence all stages of research [112]. Reflexivity is the practice of researchers critiquing how their positionality influences the research process and results [113]. Some funding agencies encourage proposal teams with members that have relevant lived experience to a research topic and/or members based in countries where research will take place [114, 115]. While not yet an explicit requirement for many funding mechanisms, reflecting on positionality and reflexivity within a team is an important exercise to identify how research team members’ identities potentially influence their relationship to the project. When developing positionality/reflexivity statements for proposals and peer-reviewed publications, we encourage researchers to use existing guides to facilitate reflection [61, 116–118].

Discussion

These tips address two aims: advancing good research integrating HCD and IS and helping teams secure the funding to pursue it. HCD and IS offer complementary strengths to health services research, although their different orientations can create tensions that researchers need to address when planning and conducting projects. Integration of HCD and IS facilitates being able to understand not only whether an EBP is usable (HCD), but what it takes to get to scale (IS). While IS excels at identifying barriers to adoption, with fidelity, of an EBP, HCD approaches can help design interventions and implementation strategies that are useable and engage clinicians and patients.

Existing IS proposal writing guides are grounded in expectations from NIH [18, 21], a major funder of IS research. While oriented towards NIH-funded IS research, these guides are valuable to those seeking to write competitive IS proposals. We build on these guides to help address additional considerations when integrating HCD and IS. Proctor et al. and Brownson et al. emphasize projects centered around evidence-based interventions, compelling research questions, appropriate models, methods and measures, and a skilled team that can carry out the work (Table 8). These characteristics are all relevant for HCD and IS research. This article focuses on additional HCD considerations in study design and proposal writing, with the expectation that researchers, manuscript reviewers, and members of grant review committees consult existing IS proposal guidelines, in addition to the recommendations offered in this paper. UWAC pilot study funding mechanism require investigators to submit an initial proposal summary and consult with UWAC faculty–who have experience working at the intersection of HCD and IS–before submitting a full proposal (Supplemental File 3). UWAC faculty also participate in proposal review. Ensuring that HCD and IS expertise is present among reviewers has been instrumental in helping applicants develop quality HCD and IS proposals and should be considered by funders seeking to support these projects.

Table 8.

Summary of recommendations for integrating HCD and IS for projects and proposals with existing IS proposal guidance

HCD and IS tips presented in this article	Proctor et al.’s 10 ingredients for IS proposals	Brownson et al.’s 12 core competencies for Dissemination & Implementation (D&I) grant writing
1. What roles do HCD and IS have in your project? 2. What are your design problems and questions? 3. What methods align with your questions and context? 4. How will you plan for uncertainty and communicate your plan? 5. What type of expertise do you need for a strong team? 6. What is your plan for engaging with users? 7. How will you measure success? 8. What safeguards and accountability mechanisms will you include to catch unintended harms as your project unfolds?	1. The care gap or quality gap 2. The evidence-based treatment to be implemented 3. Conceptual model and theoretical justification 4. Stakeholder priorities, engagement in change 5. Setting’s readiness to adopt new services/treatments/programs 6. Implementation strategy/process 7. Team experience with the setting, treatment, implementation process 8. Feasibility of proposed research design and methods 9. Measurement and analysis section 10. Policy/funding environment; leverage or support for sustaining change	Aims 1. Create a clear, rationale and realistic action plan for transforming research questions on D&I into grant proposal aims. Significance 2. Identify how to pose an innovative and significant D&I research question, justify its importance, describe the knowledge gap it addresses, and when possible, connect it to priorities of the funding agency. 3. Describe how to ground the proposal in an important quality gap that is addressable through the D&I of an evidence-based intervention, program or policy. Innovation 4. Articulate how to identify products from the D&I study, including implementation toolkits, to guide practice and policy. 5. Report on the consistency of the proposed practice change (to be addressed in the study) with the policy trends and priorities. Approach 6. Utilize an appropriate D&I model or framework to organize a proposal and integrate research questions with clear and measurable study objectives; aims; measures, and analysis strategies. 7. Explain how to document or propose measurement of the setting’s need, appropriateness and readiness of the practice change required through the D&I strategies addressed in the proposal. 8. Identify measures that clearly assess the constructs of interest in the proposed study and are practical to apply in the proposed settings. 9. Identify how to build a team with the expertise and experience for the proposed research, including D&I expertise and stakeholder experience. 10. Create a strategic dissemination plan for various target audiences that goes beyond the traditional publications and presentation at meetings. 11. Develop an analysis plan that addresses each specific aim and hypothesis and considers the different levels of analyses. Human Subjects 12. Describe the ethical (human subjects) issues that are particular to and relevant for D&I research.

Our recommendations have limitations. They are grounded primarily in our experiences with US-based teams pursuing NIH and other federal funding in mental health and behavioral health, which may limit their generalizability to other funding contexts, health sectors, or international settings. Although one author draws on projects in East Africa, examples throughout skew toward NIH mechanisms and US institutional structures. Additionally, these tips reflect our collective experience rather than systematic empirical analysis of proposal outcomes, and they should be interpreted accordingly. We nonetheless believe these tips offer a useful starting point for researchers seeking to integrate HCD and IS across settings and funders, and we encourage others to build on and refine them.

Conclusion

Developing quality projects and proposals to conduct research integrating HCD and IS requires understanding how these approaches are complementary and can result in projects that are greater than the sum of their parts. Drawing upon our own experiences being part of teams and supporting other teams carrying out health projects that integrate HCD and IS, we identify tips to help teams develop well-designed projects and proposals. We invite other researchers to share their experiences and, in alignment with HCD principles, to continue to iterate on these recommendations.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CFIR

Consolidated Framework for Implementation Research

D&I

Dissemination and Implementation

EBPs

Evidence-based programs, practices, and policies

FRAME

Framework for Reporting Adaptations and Modifications-Expanded

HCD

Human-centered design

HCI

Human-computer interaction

IS

Implementation science

NIH

National Institutes of Health

PRISM

Practical, Robust Implementation and Sustainability Model

RE-AIM

Reach, Effectiveness, Adoption, Implementation, and Maintenance

TF-CBT

Trauma-Focused Cognitive Behavioral Therapy

UWAC

University of Washington Advanced Laboratories for Accelerating the Reach and Impact of Treatments for Youth and Adults with Mental Illness Center

UX

User experience

Author contributions

TA developed the idea for the paper and co-led the writing with SAM. KO, AKG, EKP and ARL contributed to development, drafting, or review of the paper.

Funding

This paper was supported by grants P50 MH115837 and R34 MH128387, awarded by the National Institute of Mental Health, grant R305A230391, funded by the Institute for Education Sciences, and the University of Washington Population Health Initiative. The content is solely the responsibility of the authors and does not necessarily represent the official views of funders.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

N/A.

Consent for publication

N/A.

Competing interests

The authors declare no competing interests.