Defining and Evaluating In-Home Drug Disposal Systems For Opioid Analgesics: Proceedings of a Workshop (2024)
Chapter: 3 Envisioning Disposal Systems to Remove Opioids from the Home
3
Envisioning Disposal Systems to Remove Opioids from the Home
Highlights of Key Points Made by Individual Speakers*
- In-home opioid disposal systems should be easy to use and provided in a timely manner. (Agarwal, Huang)
- A number of variables can impact how people make decisions when they interact with packaged products. A human–packaging interaction model could help identify which variables are meaningful when it comes to the design and use of in-home opioid drug disposal systems. (Bix)
- Timely reminders of the need to dispose of unused opioids are needed, and approaches such as text-based reminders for patients are being studied. (Agarwal, Egan)
- There is a lack of public awareness about the risk of keeping unused opioids in the home and how to properly dispose of these drug products. (Agarwal, Egan)
- Those who may be aware of the need to dispose of unused opioids may be thwarted by a lack of useful or accurate information about available disposal options, and confusion about which option to choose. (Baran, Egan)
- Barriers to the use of opioid disposal systems can include complexity of use, cost, and a desire to hold on to unused drug. (Agarwal, Baran, Egan)
- Implementation of opioid disposal systems is impacted by culture at the patient, provider, health system, and community levels. (Huang)
* This list is the rapporteurs’ summary of points made by the individual speakers identified, and the statements have not been endorsed or verified by the National Academies of Sciences, Engineering, and Medicine. They are not intended to reflect a consensus among workshop participants.
DESIGN AND ITS RELATION TO HUMAN BEHAVIOR
Laura Bix, assistant dean for teaching, learning, and academic analytics at Michigan State University College of Agriculture & Natural Resources at Michigan State University, has worked to adapt and combine design models from different fields to manage the multifactorial issues that are involved with human–product interactions. Although her research and regulatory attention are largely focused on how human–package interactions impact behaviors such as medication selection, adherence, or compliance, she said the model could likely be applied to the interactions between humans and drug disposal systems.
Research is available on the extent to which a package may help ensure the safety and efficacy of a given product throughout its defined shelf life. Less attention has been given to the human component, specifically how people interact with a package and how this may impact their decisions and behaviors. Early studies on user-centered design were more qualitative, such as research on user comprehension of labeling. Bix established the Packaging HUB (Healthcare, Universal Design, and Biomechanics) at Michigan State to focus on quantifying the interactions between people and packaging and applying that knowledge to inform policy and package design, with the goal of improving health outcomes.
Five Stages of Information Processing
“For information to be effective, the user must go through five different stages of processing,” Bix explained (de la Fuente and Bix, 2011):
- Exposure to the information.
- Perception of the information through one of the five senses. She noted that interactions with a package or product primarily involve vision.
- Encodation of the perceived information by the brain into an electrical signal it can interpret.
- Comprehension of the encoded information (e.g., if the information encoded is that the product contains legumes, does a user who has a peanut allergy understand that the product could contain peanuts?).
- Execution of a suitable action or behavior based on comprehension of the information.
Bix then noted that the ability to process information provided on a packaged product through the five stages depends on the
- person (processing is impacted by experiences, cognitive and physical abilities, beliefs, habits, etc.);
- context (environment where the interaction with the information occurs, including distractions and competing tasks);
- task (which in the case of a drug could be proper opening, administration, storage, and/or disposal of the packaged drug product); and
- design of the packaged product.
Quantifying Consumer Interactions with Product Packaging
Studies by Bix and colleagues offer methods and lessons potentially applicable to the study of patient interaction with opioid disposal systems. One study discussed by Bix evaluated the extent to which over-the-counter (OTC) drug labeling about tamper-evident features and the lack of child-resistant packaging was prominent and conspicuous, as required by law (Bix et al., 2009). The eye movements of study participants were observed as they considered five different OTCs for purchase to determine how they interacted with five key pieces of information: “brand name, indications, drug facts label, child-resistant warning which indicates not for households with young children, and the warning alerting to tamper-evident features,” she said. One finding she highlighted was that, across all five OTC packages, more than 80 percent of participants did not look at the warning regarding tamper-evident features. She pointed out that any efforts to ensure the tamper-evident messaging on the product label was understandable were irrelevant because most people were not looking at that warning.
In another study, participants considered whether mock brands of OTC drugs were appropriate for them based on how the (fake) brand name, active ingredient, and indications were displayed on the package (Liu, 2016). Of 82 participants, 8 percent looked at the drug facts label for
every product package when deciding if they would purchase the product or not, while 62 percent never looked at the drug facts label on any of the packages.
A study of tamper-evident solutions for a food puree found that consumers could not be relied on to notice tampering or to act appropriately when they did notice tampering. The “consumers” were experts in biosystems engineering and packaging at Michigan State. When presented with the packaged puree that had been tampered with, the expert consumers were less likely to find the tampering if it was not near the container closure. Furthermore, all the experts reported having used products in their personal lives that had “a failed induction seal or a failed shrink sleeve over a product,” Bix said. Some reported not wanting to make another trip to the store to try to exchange; others reasoned that “machinery is inconsistent,” “seals pop off,” or “things happen in transit.” The proposal stemming from this study was to design a barcode with color-changing ink that could respond to “changes in the headspace gas” and make the barcode unreadable. This approach would prevent the consumer from buying the tampered product. Bix suggested that a similar approach might be considered for opioids, perhaps some process that might “slowly render the product inert.”
User-Centered Design
Numerous independent variables can potentially impact how people make decisions when they interact with packaged products. The model discussed can help to identify and assess which variables are most meaningful to study and target for intervention. Bix emphasized the importance of understanding where a system design is failing, and developing new solutions that address root causes (e.g., is it a perception problem, a comprehension problem, an execution problem?).
“Even when you create a design that is successful through processing, consumers/patients may behave in ways incongruent to what you desire/intend,” Bix concluded, and she suggested considering designs that do not depend on consumer behavior.
DESIGNING USABLE IN-HOME OPIOID DISPOSAL SYSTEMS
The International Organization for Standardization definition of usability considers the effectiveness, efficiency, and satisfaction a system affords the user, Bix said. Following the presentation, Bix was joined by four panelists for a discussion of characteristics of usable in-home opioid disposal systems, barriers to their implementation and use, and health equity considerations in the development of in-home disposal sys-
tems. Panelists included Anish Agarwal, assistant professor of emergency medicine at the University of Pennsylvania Perelman School of Medicine; Lindsay Baran, senior research director in the Health Care Evaluation department at NORC at the University of Chicago; Kathleen Egan, assistant professor in the Department of Health Education and Promotion at East Carolina University; and Lyen Huang, assistant professor of surgery at the University of Utah Spencer Fox Eccles School of Medicine. The discussion was moderated by Ruchi Fitzgerald, assistant professor at Rush University and service chief of inpatient addiction medicine at PCC Community Wellness Center.
Characteristics of Usable In-Home Opioid Disposal Systems
From his perspective as an emergency medicine clinician and health care innovator, Agarwal said in-home opioid disposal systems need to be easy to use and provided in a time-sensitive manner. Agarwal described a study in which patients are engaged via text messaging during the week following their surgery (Agarwal et al., 2021). Participants report their use of the prescribed opioid and their pain score and are reminded “in rapid, real-time fashion” when it is appropriate to dispose of unused drug. Agarwal then described a follow-up randomized controlled trial (RCT) in which, in addition to the text messaging, in-home disposal systems were mailed to participants to arrive when it was predicted they would no longer be taking the prescribed opioids (most frequently timed to arrive 4 to 5 days postoperatively) (Agarwal et al., 2022). Feedback from patients indicated that they receive a large packet of information upon discharge and often forget about any information on the need to dispose of unused opioids at some future time. The study found that patients who were postoperatively mailed easy-to-use disposal kits and received text message reminders to dispose were “twice as likely to dispose,” he said.
Egan agreed with the need for more timely disposal messaging and said she is also studying a text message intervention that reminds patients to dispose as they near the end of their prescription (Egan et al., 2020). A challenge, she noted, is the wide variability in prescription length. Huang also emphasized the importance of the timing of interventions. University of Utah Health provides free disposal kits to all patients who fill a benzodiazepine or opioid prescription. Huang’s studies of real-world use of in-home opioid disposal kits have found that there is intent to dispose but not follow-through (Huang et al., 2023). Postoperative follow-up surveys indicated that 51 percent of patients want to or plan to dispose, but at 2 weeks postsurgery only 26 percent have actually done so. Huang highlighted the need to address patient-level barriers to disposal and to better target messaging.
Barriers to Safe Disposal of Unused Opioids
Egan shared lessons from her experience working with community coalitions to address substance use and expand disposal opportunities. She said recurrent feedback from community members is that there is a lack of awareness and education about the need to dispose of opioids and how they should be disposed (including the use of disposal systems). Agarwal also emphasized the need to communicate the risk that opioids present for others in the patient’s home and in the community, and the importance of disposing of unused drug.
People who recognize the risk of having opioids in the home “are more likely to be interested in disposal and actually dispose,” Egan said. There are concerns, however, about the safety of the disposal systems themselves, and the potential danger to children or animals that encounter or ingest the systems. In addition, those who are aware of the need to dispose are often “immobilized by all the options.” For those who decide to go to a take-back location, for example, the DEA list of public disposal boxes is often inaccurate. She explained that pharmacies must register to have a disposal box, but they might not install a box or might remove an existing box. Community feedback suggests that people will try to dispose at a pharmacy one time. If there is no disposal box or the box is closed, they are unmotivated to try again, she said. Baran agreed with Egan that lack of awareness is a main barrier, and that many people not only are unaware of existing drug disposal options but also struggle to find accurate information. An Internet search for “how to get rid of unneeded opioids” nets wide-ranging results. There is valuable information from reputable sources such as federal websites about disposal options, Baran said, “but for the average person who is just looking for a clear answer about how to get rid of their medication, it’s incredibly complicated to navigate.”
Barriers to using some of the opioid disposal options that are available include complexity of the disposal system, costs, and in some cases, the need to travel to a disposal location, which Baran noted requires transportation and has costs. Egan said there is feedback from communities that the cost of opioid disposal systems is a significant barrier to uptake. While there might be interest in having a disposal kit, people are generally not willing to pay for it, she said. Egan also noted that kits might not be used to dispose of opioids. She described one intervention in which community members were provided with a packaged disposal system (in this case, the Deterra system). Most found the disposal system easy to use, but further analysis revealed that many intended to use the system to dispose of drugs other than opioids.
Another barrier to disposal of unused opioids is that, for a variety of reasons, some people do not want to dispose of them, Baran said. She
reiterated the point by Robert Hoffman that people are often unwilling to give up something they paid for, even if they no longer need it, because they believe they might need it in the future. In this regard, barriers to accessing prescription opioid medications for pain management are a barrier to disposal. Agarwal said that when informed about the need for disposal, his patients often raise concerns about possibly needing their pain medication in the future. Baran said people, especially those with episodic pain conditions, often have “a justified fear” of not being able to obtain additional medication if needed.
Huang said that when envisioning in-home disposal of opioids, it is important to acknowledge the culture at the patient, provider, health system, and community levels in which implementation is being attempted. For example, he said that algorithms and electronic health record (EHR) prompts designed to help clinicians reduce opioid prescribing have been implemented, “but the most challenging part has been convincing attending surgeons to go along with this low- or no-opioid prescribing.” Medical residents have reported wanting to avoid prescribing opioids at discharge when the patient has not taken any for 48 hours, but they often prescribe anyway out of concern that the patient will call later with pain or the attending will be displeased that nothing was prescribed. Convincing patients they do not need opioids “just in case” is another cultural barrier to both prescribing and disposal.
In addressing barriers, Bix said it is important to identify which behaviors need to be motivated and in what situations, understanding that “people are incredibly unpredictable.” She reiterated the importance of empirical evaluation of the system to identify where it is failing and to address the root causes. Huang said research is needed on which types of interventions can successfully motivate the desired opioid disposal behaviors.
Targeting In-Home Disposal Interventions
Panelists discussed whether to focus on particular subgroups for in-home disposal interventions. Egan said that patients treating chronic pain tend to use up their medication within the prescribed period, and harm reduction strategies discussed by Hoffman, such as secure storage, are needed. Baran proposed thinking about subgroups based on prescribing practice (e.g., short term versus long term) rather than on condition (e.g., chronic versus acute). Chronic and acute pain are not distinct subgroups and there are overlaps. As an example, she shared that she personally has a chronic pain condition but does not use opioids on a chronic basis. An opioid might be prescribed short term, for example, if she were to have surgery to address an acute flare of her condition. Baran also suggested
that in-home disposal is a “universal prevention intervention,” but outcomes and measures of success might be population specific. For example, a reduction in child poisonings might be one measure of success of the uptake of in-home disposal of opioids by parents.
Agarwal highlighted the need to understand the “blind spots” in prescribing. He observed that patients are often prescribed a particular dose and duration of opioid for an injury or after surgery because that is the dose and duration that has always been used. It is not possible to know who will need more pain management after a procedure and who might need none. But prescribers want to know when their patients are in pain, Agarwal said, and his group is studying a real-time communication mechanism to report pain management concerns from patients to their surgeons and other prescribers. This type of communication is a change in the culture of prescribing, he said. Furthermore, “surgeons are profoundly trusted resources for patients,” and can inform patients about the risks of opioids in the home and encourage them to dispose of unused medication using a provided in-home disposal kit. He discussed developing a learning health system to inform prescribing guidelines that better anticipate patient pain management needs and tailor prescribing accordingly, and provide timely information, reminders, and supplies for disposal.
Rather than focusing on particular subgroups for disposal interventions, Huang suggested taking a patient-centered approach to disposal. The capability exists to collect feedback from individual patients about real and perceived barriers to their disposal of their unused opioids and provide “very targeted interventions that address those concerns,” he said. In essence, this is “disposal intervention as personalized medicine,” he added.
Health Equity Considerations
R. Fitzgerald raised the issue of health equity in the development and implementation of in-home opioid disposal systems. Key considerations for equitable access to disposal systems include the cost and ease of acquiring the disposal system, Baran said. For example, is the disposal kit provided when the medication is dispensed, or does the patient have to travel again or elsewhere to acquire it or pay to have it shipped? Are additional supplies required? Raising awareness of the need to dispose of unused opioids and of the options available is also a component of ensuring equitable access. As discussed, ease of use is an important feature of any disposal system for use by the public, but this feature might be especially important for individuals who are homeless or not in a traditional home setting. The systems, including the instructions, should be both cognitively and physically accessible, Baran continued. For example,
instructions should be clear, and the process should not require many steps or take much dexterity or manual effort (e.g., mixing). She suggested that having support available from a pharmacist, prescribing physician, or perhaps a toll-free number would also be helpful. In summary, Baran said systems should be “easy to access; free or low-cost … easy to explain and easy to use with little room for error;” and safe for users, children, pets, and the environment. Egan shared feedback from the communities she works with that many find the instructions for disposing of opioids challenging to follow. She added that accessibility includes making the font large enough for older users to see.
Bix suggested that the efficiency component of usability is a key measure for designing equitable and accessible in-home opioid disposal systems. Efficiency addresses whether the user has the resources needed to realize the outcome the product is intended to facilitate, she said. These resources could be physical strength, cognition, ability to follow the steps required, ability to travel, and access to transportation, for example. Different populations might face different usability challenges. In designing a product, it is important to understand at the outset which population the product is intended for and what behavior it is intended to motivate, Bix said.
Egan reported that implementation of disposal kiosks at pharmacies throughout communities in North Carolina was equitable to the extent that it did not vary by the zip code of the pharmacies. Variability in implementation of disposal kiosks was associated with the type of pharmacy, due to factors that Egan said could also translate to variability in the dispensing of in-home disposal kits. Pharmacists in corporate-owned pharmacies said cost is not a factor in the implementation disposal kiosks, but the managing pharmacist does not have control over whether a kiosk is installed in their pharmacy. Independent pharmacists reported that cost is a barrier, and installation of a disposal kiosk is often supported by external funding. Huang agreed there can be challenges of providing resources to patients through pharmacies and said pathways are needed to ensure that distribution is not limited to large-chain pharmacies or large health systems for whom cost is not a barrier. Egan added that community substance use coalitions are using funding from the ONDCP Drug Free Communities program to distribute in-home disposal kits, but there are not yet outcomes data to assess whether this approach is making an impact.
Agarwal pointed out that unequal access to digital technology and digital health could impact equitable implementation of disposal kits. He suggested looking for opportunities to place disposal resources in accessible places (as has been done with the placement of Narcan in libraries in Philadelphia and needle drop boxes in airport restrooms).
Huang said rurality impacts access to health care providers, pharmacies, and disposal kiosks. He pointed out that “rural” is subjective. In the mountain west where he lives, patients often travel 5 or 6 hours for health care or more than an hour to the nearest pharmacy, which can be “over a mountain range that’s inaccessible in the middle of wintertime because of the wind and snow.” Huang said the timed mailing of disposal kits, as done in the study by Agarwal, could be a challenge in rural areas for patients who do not have a traditional mailing address (as is the case on some Tribal reservations). Rurality also impacts pain management, he said. While a provider can promise to remotely prescribe a new or refill prescription when the patient needs it, the patient might not be able to go to the pharmacy to pick it up.
Lessons from Naloxone Distribution
Agarwal drew a parallel between distribution of naloxone and distribution of in-home disposal systems in that the goal of both is to provide something to the patient that will not be used right away. He conducted a study which found that most patients who were dispensed naloxone spray at discharge from the emergency department were carrying it with them in the week after discharge. By contrast, most patients who were discharged with a prescription for naloxone were not carrying it, and the need to go and fill the prescription seemed to be a barrier. Egan shared feedback from a pharmacist who reported that even a small copay of $4 dissuaded people from filling a prescription for naloxone. She suggested that patients would also be unlikely to accept a copay for an in-home disposal kit. Huang described efforts to increase distribution of naloxone across Utah. A challenge has been that patients “overwhelmingly prefer the nasal spray version versus injection form,” he said, but the spray can cost patients hundreds of dollars out of pocket, and many do not pick it up from the pharmacy. R. Fitzgerald noted that the availability of naloxone varies by state and locality and is available to everyone free of cost in Illinois.
