Take-Home Naloxone Kits: Attitudes and Likelihood-Of-Use Outcomes from a European Survey of Potential Overdose Witnesses

Background: Injectable naloxone is already provided as take-home naloxone (THN), and new concentrated intranasal naloxone is now being introduced in Europe. Despite evidence of the effectiveness and cost-effectiveness of THN, little is known about the attitudes of key target populations: people who use opioids (PWUO), family/friends, and staff. We examined the acceptability of different naloxone devices (ampoule, prefilled syringe, and concentrated nasal spray) across 5 European countries. Objectives: The aim of this study was to compare THN target groups (PWUO vs. family/friends vs. staff) in their past rates of witnessed overdose and THN administration (as indicators of future use), current THN device preference, and THN carriage on the day of survey. Method: Cross-sectional survey of respondents (age ≥18) in addiction treatment, harm reduction, and recovery services in Denmark, England, Estonia, Norway, and Scotland. A purpose-developed questionnaire (59 items) was administered in the local language electronically or in a pen-and-paper format. Results: Among n = 725 participants, 458 were PWUO (63.2%), 214 staff (29.5%), and 53 (7.3%) family members. The groups differed significantly in their likelihood-of-future THN use (p < 0.001): PWUO had the highest rate of previously witnessing overdoses (352; 77.7%), and staff members reported the highest past naloxone use (62; 30.1%). Across all groups, most respondents (503; 72.4%) perceived the nasal spray device to be the easiest to use. Most reported willingness to use the spray in an overdose emergency (508; 73.5%), followed by the prefilled syringe (457; 66.2%) and ampoules (64; 38.2%). Average THN carriage was 18.6%, ranging from 17.4% (PWUO) to 29.6% (family members). Conclusion: Respondents considered the concentrated naloxone nasal spray the easiest device to use. Still, most expressed willingness to use the nasal spray as well as the prefilled syringe in an overdose emergency. Carriage rates were generally low, with fewer than 1 in 5 respondents carrying their THN kit on the day of the survey.


Introduction
Injectable naloxone is already provided as take-home naloxone (THN) in Europe, and licensed nasal spray products ("concentrated naloxone nasal spray") have become available since 2018 [1,2]. There is strong evidence for the effectiveness and cost-effectiveness of THN presupply to potential overdose witnesses, i.e., people who use opioids (PWUO), their family/friends, and staff in contact with PWUO [3][4][5]. Yet, little is known about the target groups' attitudes toward different THN devices or their likelihood to witness overdoses. Data from Scotland [6,7] and the USA [8] show low THN carriage rates (5-16%) among PWUO. This presents a central implementation challenge to community-based naloxone access in overdose emergencies, potentially reducing benefit from THN.
We thus surveyed potential overdose witnesses about their current attitudes and past experiences. We aimed to compare THN target groups (PWUO vs. family/friends vs. staff) in their: 1. Likelihood to use THN in future overdose emergencies. On the grounds that future behavior is best predicted by past behavior, we defined this as rates of past (a) witnessed overdoses and (b) THN administration 2. THN device preference (ampoules vs. concentrated nasal spray vs. prefilled syringe), assessed as (a) perceived ease of use and (b) willingness to use in an overdose emergency 3. THN carriage rates

Materials and Methods
Eligibility Adults (≥18 years) were asked to participate if they were (1) PWUO (current and former users, any treatment status), (2) their family members, or (3) staff working with PWUO. Individuals who could not effectively communicate with the interviewer due to intoxication or language comprehension barriers were excluded.

Recruitment
Data collection occurred in 8 cities (see Table 1) across Denmark, England, Estonia, Norway, and Scotland via convenience sampling between April 2018 and June 2019, with PWUO and family being recruited from addiction treatment, harm reduction, and recovery services and staff from staff meetings. All were given a participant information sheet, alongside verbal summary, and could clarify questions with the interviewer before providing informed verbal (London sites) or written consent (all other sites).

Reimbursement
Only the Glasgow and Birmingham sites offered participating PWUO and the family a 5 GBP shopping voucher as compensation for their time.

Research Ethics
Ethics advice was sought locally. The Birmingham and Solihull Mental Health NHS Foundation Trust provided local approval (reference number: SE0123). All other sites waived the requirement for ethics review due to anonymized data collection (Copenhagen), recruitment outside NHS services (Glasgow), or the survey being conducted as quality improvement project (London, Estonia, Norway).

Measure
Interviewers (i.e., researchers or keyworkers) administered the purpose-developed questionnaire (translated from English into local languages for Denmark, Estonia, and Norway) in an iPad-based or pen-and-paper format. All responses were anonymous. The questionnaire took 10-15 min to complete and comprised up to 59 items (with skip patterns) covering the outcomes: history of witnessed overdose (yes/no), overdose rates (continuous), naloxone use (yes/no), THN daily carriage (yes/no), and THN device preference (see below).

Analysis
The data were analyzed in SPSS Statistics 25.0 (IBM, Armonk, NY, USA). The χ 2 test was conducted for categorical variables, and ANOVA was performed to compare group counts of witnessed overdoses. A p value below 0.05 was considered statistically significant. For measures of THN device preference, no inference statistics were calculated since responses were not mutually exclusive (e.g., "Which one(s) would you be willing to use in an overdose emergency"?).

Results
In the following sections, we report sample demographics and outcomes (see also Tables 1,2

THN Device Preference
Willingness to Use in OD Emergency Across groups, most respondents (508; 73.5%) were willing to use the nasal spray in an overdose, followed by willingness to use the prefilled syringe (457; 66.2%) and ampoules (64; 38.2.%). Nasal spray preference was most pronounced among staff (164; 81.6%). The smallest difference in preference ratings was observed among PWUO, with little difference between the nasal spray (69.3%) and prefilled syringe (64.5%).

THN Carriage Rates
No significant association was found between group and THN carriage (χ 2 (1, N = 297) = 1.496, p = 0.221). Among those supplied with THN, only few PWUO (45; 17.4%) and family members (8; 29.6%) reported carrying their kit on the day of the survey -equivalent to 18.6% (53/285) across both groups. Staff were surveyed during working hours and not asked about carriage.

Discussion
Our survey of potential overdose witnesses in 5 European countries has yielded 3 central findings. First, groups differed significantly in their experience of witnessing overdoses and using THN. Using past behavior as a predictor for future behavior, we found that PWUO had the highest likelihood of witnessing future overdoses, with >3 quarters having previously been present at an opioid overdose. Almost one-third of staff members reported past naloxone administration (i.e., the highest rate across groups). These findings suggest that PWUO and staff should be the primary target groups for THN distribution.
Second, THN device preference was less pronounced than expected. In an earlier Australian survey [9] which preceded the development of a concentrated naloxone nasal spray, nearly 3 quarters of participating heroin users (n = 99) had preferred nasal spray to injection. Similarly, in our survey, almost 3 quarters of all respondents considered the nasal spray easiest to use and reported they would administer the spray in an overdose emergency. Nevertheless, over two-thirds of respondents stated that they would be willing to use the prefilled syringe in an emergency. These results highlight that both the concentrated nasal spray and prefilled syringe would seem suitable for layperson distribution. Nasal spray preference was more pronounced among family members and staff (see Table 2), suggesting that these groups may be more receptive to noninjectable devices -likely because they have less experience with injectable devices than PWUO.
Third, carriage rates were generally low, independent of the different THN supply rates (36-93%) in the survey countries (see online supplementary material). Less than a fifth of PWUO and family members (19%) had their THN kit available on the day of the survey, consistent with 5-16% carriage rates previously reported among PWUO in Scotland and the USA [5][6][7]. While local supply rates likely reflect the degree of THN implementation (incl. funding), the reasons for low THN carriage rates are poorly understood. Our findings highlight the need for emphasizing the importance of carriage during THN training to ensure naloxone access in overdose emergencies.
Our study is the first to assess naloxone device preference and likelihood to witness overdoses across different THN target populations. Notably, we managed to recruit over 700 respondents across 5 European countries, which strengthen the external validity of our findings. However, due to their limited presence at recruitment sites, family members (n = 53) were underrepresented in our sample, and the generalizability of their responses is unclear. Our use of convenience sampling may have produced a nonrepresentative sample, with overrepresentation of male PWUO (333; 72.7%).
Further limitations include our cross-sectional design involving self-report data, which is inherently prone to social desirability and recall biases. Moreover, our survey was conducted before a concentrated nasal spray became commercially available in 3 survey countries (in Denmark and Norway, THN programs were already distributing concentrated nasal sprays, and an earlier device, i.e., 2 mg/2 mL multistep nasal sprays, was still in circulation but not included in the survey). Questions about device preference were thus mostly hypothetical and did not ad-DOI: 10.1159/000521197 dress dosage or cost. Preliminary data on the prehospital use of a different concentrated nasal spray (i.e., a 4 mg/0.1 mL product) are available from the USA, reporting on 261 overdose cases [10]. Future research should use prospective designs to recruit cohorts of potential overdose witnesses who are provided different THN devices (e.g., concentrated nasal spray vs. prefilled syringe) and compare their carriage rates and experiences of THN use as well as post-overdose behavior (e.g., help-/treatment seeking).

Conclusion
The concentrated naloxone nasal spray was generally preferred, but prefilled syringes were also considered acceptable for use in an overdose emergency, especially among PWUO. Nasal spray may be particularly suitable for distribution to staff and family members, alongside tailored training.
PWUO were most likely to witness overdoses, but <1 in 5 were carrying their THN kit on the day of the survey. This highlights the need to promote naloxone carriage to maximize naloxone access and prevent overdose mortality.