Design details for overdose education and take‐home naloxone kits: Codesign with family medicine, emergency department, addictions medicine and community

Abstract Introduction Overdose education and naloxone distribution (OEND) programmes equip and train people who are likely to witness an opioid overdose to respond with effective first aid interventions. Despite OEND expansion across North America, overdose rates are increasing, raising questions about how to improve OEND programmes. We conducted an iterative series of codesign stakeholder workshops to develop a prototype for take‐home naloxone (THN)‐kit (i.e., two doses of intranasal naloxone and training on how to administer it). Methods We recruited people who use opioids, frontline healthcare providers and public health representatives to participate in codesign workshops covering questions related to THN‐kit prototypes, training on how to use it, and implementation, including refinement of design artefacts using personas and journey maps. Completed over 9 months, the workshops were audio‐recorded and transcribed with visible results of the workshops (i.e., sticky notes, sketches) archived. We used thematic analyses of these materials to identify design requirements for THN‐kits and training. Results We facilitated 13 codesign workshops to identify and address gaps in existing opioid overdose education training and THN‐kits and emphasize timely response and stigma in future THN‐kit design. Using an iterative process, we created 15 prototypes, 3 candidate prototypes and a final prototype THN‐kit from the synthesis of the codesign workshops. Conclusion The final prototype is available for a variety of implementation and evaluation processes. The THN‐kit offers an integrated solution combining ultra‐brief training animation and physical packaging of nasal naloxone to be distributed in family practice clinics, emergency departments, addiction medicine clinics and community settings. Patient or Public Contribution The codesign process was deliberately structured to involve community members (the public), with multiple opportunities for public contribution. In addition, patient/public participation was a principle for the management and structuring of the research team.


| INTRODUCTION
Overdose education and naloxone distribution (OEND) programmes are intended to both widen access to the opioid antagonist naloxone and enable effective first aid response. 1,2 However, the number of overdose-related deaths continues to increase, which brings into focus the design of OEND programmes, including the design of take-home naloxone (THN) kits, the training to use them and programmes to distribute them. 3 The rapid rate of implementation of a variety of OEND and THN-kit programmes has enabled naloxone to be distributed quickly, but there is a gap in knowledge about the details of the THNkits that are being distributed and the training tools to use them, including the packaging, the training topics and format and the access to and awareness of THN-kits and training at distribution sites. [3][4][5][6] Many THN-kits contain instructions detailing the steps to follow in using naloxone; however, we are aware of only one study evaluating a THNkit instruction handout. This study by the Food and Drug Administration assessed a product label for naloxone only 7 ; it did not cover training or the experience of accessing, stowing, sharing or using a THN-kit.
The components that support THN-kit distribution, training and use include awareness tools, media assets, training tools (such as instructions), THN-kit packaging, and sharing and access mechanisms.
Increased attention to the details of THN-kit design is needed to meet the increased need to make THN-kits available to lay responders who may or may not have experience with naloxone administration. 7 Just one example of a design aspect in need of attention is the packaging.
Recent research indicates a need to examine the design of carrying cases for naloxone THN-kits due to the negative attention they receive as a result of the stigma associated with drug use. 8 Naloxone packaging has three functions: it must address stigma and establish an identity and meaning through form and visual design; it must protect the contents, and it must be practical to use (including carrying, opening, closing and accessing easy-to-understand instructions to reduce memory load and task complexity). 9 Recognizing the need to integrate evidence-informed knowledge of the design of THN-kits and training with insights from people with lived experience of opioid use and overdose, who understand stigma firsthand, we took a participatory and codesign approach to the design of a THN-kit and training 10 for distribution in family medicine clinics, addictions medicine clinics and emergency departments for lay use. 11 Before conducting the research reported in this paper, we held a multistakeholder workshop to elicit design considerations for OEND programmes more generally. The multistakeholder workshop addressed issues of stigma and marginalization 12 and resulted in SELLEN ET AL. | 2441 seven considerations for the design, distribution and use of naloxone training and THN-kits. These considerations served as a starting point for the development of basic prototypes with stakeholders using a codesign process. 12 In this article, we describe what we learned from an iterative series of codesign workshops with varied stakeholders to develop a prototype for a THN-kit that includes packaging for two doses of intranasal naloxone and training on how to administer it.

| Objectives
We aim to integrate evidence-informed knowledge of overdose first aid with insights from people with lived experience to design a THN-kit and training specifically for distribution in family medicine clinics, addictions medicine clinics and emergency departments for lay use.

| MATERIALS AND METHODS
This study is part of the larger Surviving Opioid Overdose with Naloxone Education and Resuscitation (SOONER) Project, which combines codesign, clinical trial and community engagement elements. The goal of the SOONER Project is to develop and evaluate an effective THN-kit and training and to reduce opioid-related stigma and inequity. The SOONER Project has three phases: Phase I was a participatory codesign initiative in which scientists, design researchers and community members cocreated a THN-kit and training that will be evaluated in subsequent phases, and is the subject of this paper. Phase II is a multimethods feasibility study for a randomized controlled trial 11 ; and Phase III is a full-scale randomized trial. 13 The SOONER Project is a collaboration between OCAD University, Unity Health, Inner City Health Associates, University of Toronto and Toronto Public Health.

| Study design
To inform the design of the THN-kit and the training tools to use it, we integrated evidence from the literature, best practices in communication design and feedback from codesign. Our study included community engagement and relationship building with people with lived experiences of drug use and overdose. This engagement involved using both participatory design methods 14,15 and more codesign techniques. 16,17 Table 1 outlines the steps in the codesign process.
Stakeholder representatives-people who use opioids, frontline healthcare providers and public health representatives-were directly involved as members of the research team in decision-making and design feedback throughout the process. In addition, a community advisory panel was developed to enable a continuous mechanism for community feedback, decision-making and involvement. This approach to participation allowed for people who use opioids (both prescribed and nonprescribed) to be involved how, when and as much as they desired. Compensation was provided on an hourly basis for all forms of engagement. As well, meetings and research activities began with a shared meal. All researchers attended cultural safety and sensitivity training before beginning research activities.
The codesign process was conducted over 9 months. It included three advisory council workshops covering training style and direction, the information in and contents of the training, the use of language and symbols, the material and colour choices for the THN-kit packaging, as well as refinement of design artefacts, including personas and journey maps. 14,15 The personas and journey maps are used to collate insights about stakeholders' roles and characteristics (one persona representing each stakeholder group) and insights from observations that are captured in a visual representation of the experience of each setting as someone who is at risk of overdose (addictions medicine, family medicine) or has experienced an overdose (emergency department). The personas and journey maps are available online as supporting material to this paper at www.soonerproject.ca.
Two codesign workshops were conducted in each setting for which the THN-kit was being designed, including family medicine, addictions medicine and emergency departments. Twenty-four guiding questions (see Figure 1 for examples) split into three themes (training, packaging, implementation) were used to structure the workshops.
Three codesign workshops were conducted in community settings for people who use opioids (both prescribed and nonprescribed

| Setting
The project setting was a Canadian urban centre with a public health response to overdose that is informed by a harm reduction approach.
Observations occurred at each of the family medicine clinics, emergency departments and addictions medicine clinics involved in the study, over

| Data collection
A variety of craft materials and resources (sticky notes, pens, stickers, prompt cards, prototypes) were supplied at the codesign workshops.
The workshops were audio-recorded, and the visible results of the workshops (i.e., sticky notes, sketches) were photographed and archived. Audio recordings of each session were transcribed verbatim.

| Analysis
The codesign workshops were structured with 24 guiding questions on specific design aspects of the THN-kit, training and implementation relating to the seven design considerations and supporting research. The subsequent qualitative approach followed Braun and Clark 18 in conducting the analysis of the transcripts initially based on these guiding questions. All codesign workshops were audiorecorded, transcribed, reviewed and then analysed. The transcripts were initially organized using the question groupings as a preexisting conceptual framework in the analytic process. 19 To analyse open-ended discussions, three researchers reviewed transcripts to identify topical categories. The second step included second and third readings and review and further grouping into broader categories, departing from the 24 questions and developing themes. We collaboratively developed themes to illuminate how stakeholder perspectives converged and diverged. Analysis was an iterative process in which themes were formed and F I G U R E 1 Codesign materials-example questions and prompt cards.
refined. 20 This grouping, review and regrouping process was carried out manually 21 with printouts of the transcripts cut into snippets and physically grouped on a large table; we then named and described each theme with its stack of quotes, photographing and keeping descriptions of each theme. We maintained links back to the codesign session and role using a colour-coding system. This manual process was chosen to enable collaborative discussion across transcripts and themes and support visual and tactile engagement as has been described by Maher et al. 22

| RESULTS
Over 70 individuals took part in the codesign workshops, with many attending more than one workshop ( Table 2).
The codesign process (see Table 2) progressed from one session to the next, with refinements made between sessions as design directions were closed off and more detailed design work was undertaken.
Feedback was continuously incorporated into the prototypes, culminating in the final coordinated THN-kit, described below.

| THN-kit design results
The final THN-kit design, resulting from the combination of existing evidence, design review and feedback from the codesign process, comprises two nasal naloxone sprays in a small resealable waterproof flexible package. The packaging is designed to be visually identified as first aid supplies, with a green/grey cross and a solid/ prominent visual style. Inside the package, the two sprays are clearly visible; the physical layout of the package supports the sequence of responses in an overdose emergency aided by simple infographics (Figure 2A). These infographics and visual style are echoed in a 2-min training animation that contains supportive, nonstigmatizing imagery and language ( Figure 2B) and is designed to improve retention and comprehension using three repetitions of the first aid steps (all steps, repeated steps and visual/word-based). The animation supports auditory as well as visual cognitive styles.

| Visual design
Background work and results from the multistakeholder workshop 12 indicated the need for a visual language for all aspects of the project that did not perpetuate stigma, had clear messaging and respected privacy (see Figure 3 for conventional designs). For the visual language for the project, we emphasized positive, supportive and energetic concepts. After reviewing common first aid symbols with the community advisory panel, including existing THN-kits and ways in which naloxone is presented, we chose to maintain the link to first aid with the first aid cross symbol but use graphic treatments to soften the symbol together with graphic elements derived from the shape of the 'spray' of the nasal naloxone (see Figure 4). Colour choice was discussed during the codesign workshops, contrasting existing colour schemes in the medical (e.g., first aid red cross) and harm reduction community (black/red, purple/violet, neon pink) to alternatives. Participants noted that while the red cross is a recognizable symbol of first aid, there are alternative colours that may also signal first aid without using 'danger' activating colours.
Most groups supported the use of a green colour palette but also supported the production of THN-kits in a variety of colours to enable choice. Our design review indicated that the colour schemes and materials of existing harm reduction THN-kits (black and white, paper/small black plastic bags, orange, black/red) were not

| Packaging design
Intranasal naloxone together with educational and other materials creates a bulky bundle in need of an external cover/container. During codesign workshops, healthcare provider participants described the existing zippered THN-kits handed out by The Works, a public health programme, as being 'really nice', though they did express concern regarding the cost. Community participants felt that some THN-kit packaging was overbuilt, which would lead people to use the containers for other purposes (to hold pens or small personal items) and described them as being 'too good' and others as 'Red Cross-like' (see Figure 1) because a hard-outline bright red cross is commonly depicted on THN-kits. In response, we prototyped a wide range of packaging using different materials and diverging form factors ( Figure 5). As the codesign process progressed, the issues of size, the form factor, the need to be able to show trusted friends/family the THNkit and reseal it, as well as material costs and printing, led us to consider waterproof sheeting as a material (see Figure 6). In the community, there were mixed opinions on its durability, with general concern about wear and tear.

| Training design
We Animated videos have been found to be effective in providing information and are typically perceived as nonthreatening, familiar, and accessible across age groups, cultures and literacy levels. 25,26 Animation may hold the attention of viewers and enhance recall, and it has been shown to be more effective than live action as an educational tool. Animation allows control over presentation, characterization, staging and timing. 25 Live action requires detailed choices about who and how to represent-in this case in a situation of overdose, which would be emotionally activating but potentially also deeply stigmatizing. The level of control that animation provides is particularly helpful when dealing with such a stigmatizing topic as drug use and overdose. While an animation approach was considered from the outset, in the early stages of codesign live action was also considered.
Feedback at this stage indicated that abstract animation would offer a nonstigmatizing training experience. The design direction at this stage was driven by issues, such as the need for an animation to depict characters experiencing overdose without discernible gender or race, to avoid stereotypical depictions of people who use drugs and where they use them (Figure 7). The team subsequently created a script for the training codesign workshops that enabled issues of potentially triggering and problematic language to be addressed early.
This codesign process together with evidence on the design of public health communication, effective emergency and first aid training materials, 26-28 guidance specific to overdose and current first aid guidelines 29-31 resulted in a set of design requirements for training (see Table 3) that informed training prototypes 1 and 2.
In reviewing research on effective first aid training, we found language around calling 911 to address users' potential reluctance to call.

| Interaction design
Several design requirements were identified from the multistakeholder workshop 12  share, refresh and use. 33 Community members supported the integration of an infographic directly inside the THN-kit packaging, so it would not be lost. Some participants also thought the infographic could be set as their phone's wallpaper or on a small wallet-sized card. By integrating these ideas into the final packaging prototypes, several design requirements were identified (Table 4) to guide these next steps ( Figure 6).
This design development was positively received by each participant group: Early storyboards for codesign of training contents and depiction of the response. and then you take out the, the that's really yeah. P2:

T A B L E 3 Training design requirements and training basic steps
Yeah, that's awesome I like that one. P: It's really cool.

| DISCUSSION
High-stress situations warrant designs that support information processing and understanding that will lead to intended action(s). The focus on simplified THN-kit contents, an ultra-brief schematic for first aid steps and visual choices that position the THN-kit as first aid address the need to make carrying naloxone a normative choice. 8 Cultural, language and contextual issues specific to overdose informed the final designs. Highlighting the need to call 911 as part of the training and emphasizing nonstigmatizing language and nonspecific settings and characters was a key strategy to enhance uptake and use. Relying on first aid symbols to identify the THN-kit as a first-aid supply was another key decision that aligns with current research, highlighting the need to address the fear of association between naloxone and drug use/criminality, 35,36 to specifically focus on how simple and effective naloxone is to use 37 and to support recognition of and carrying of naloxone among a broader range of potential lay responders and people who use drugs. Harm reduction principles aim to meet users 'where they are at' and should reflect individual and community. 42 Accepting, understanding and recognizing the reality and complexity of the many factors affecting people who use drugs is important for community engagement to ensure their experiences are represented accurately in ways that are not discriminatory or stigmatizing. Stigma is a major barrier to enabling a wider public response to overdose as a first aid emergency. Stigma was addressed at every design level-from detailed design decisions about labelling, colour choice and material choice, to packaging shape and affordances, to training language and style, format and delivery. The codesign process was deliberately structured to provide flexible opportunities for people who use drugs, potential lay responders and representatives from each of the delivery settings to shape design decisions. At numerous steps, specific details were changed considering new information and new ideas to address stigmatizing language, concepts and choices. This iterative process built trust over time that stigma-based concerns would be addressed by the design team to enable participants to feel comfortable using and sharing the THN-kits. This process has led to a design that specifically responds to stakeholder needs in an urban Canadian context; we do not know if the design would also resonate with a different setting or demographic (i.e., nonurban, suburban, specific cultural community or geographic region). We recommend additional codesign steps to test for appropriateness and necessary adaptations in different settings or with different groups.

| CONCLUSION
Addressing stigma and marginalization necessitated community engagement and relationship building with those with lived experience so that verbal and visual language issues could be addressed and programming. The intention of the THN-kit and training are that they enable anyone to be a lay responder to overdose.