“Feeling confident and equipped”: Evaluating the acceptability and efficacy of an overdose response and naloxone administration intervention to service industry employees in New York City

Abstract

Background:

The problem of injection drug use in public bathrooms has been documented from the perspectives of people who inject drugs and service industry employees (SIEs). Previous studies suggest that SIEs are unaware of how to respond to opioid overdoses, yet there are no behavioral interventions designed for SIEs to address their specific needs. In response to this gap in the field, we constructed, implemented, and evaluated a three-module behavioral intervention for SIEs grounded in the Information-Motivation-Behavioral skills model. This paper focuses on the evaluation of one module, namely, the intervention component addressing overdose response and naloxone administration (ORNA).

Methods:

Participants were SIEs (N = 18 from two separate business establishments) recruited using convenience sampling. The study utilized a pre-/post-test concurrent nested mixed method design and collected quantitative and qualitative data including an evaluation of the intervention module. The primary outcomes were opioid overdose-related knowledge and attitudes. Acceptability was also assessed.

Results:

SIEs demonstrated significant improvements (p < 0.01, Cohen’s d = 1.45) in opioid overdose-related knowledge as well as more positive opioid overdose-related attitudes (p < 0.01, Cohen’s d = 2.45) following the intervention. Participants also reported high levels of acceptability of the module and suggestions for improvement (i.e., more role-playing).

Conclusions:

This study highlights the acceptability and evidence of efficacy of the ORNA module, as well as the utility of training SIEs in ORNA. The expansion of this training to other SIEs and public employees (librarians, etc.) who manage public bathrooms warrants further investigation.

1. Introduction

Drug overdose mortality rates per year in the United States (US) nearly tripled between 1999–2014 with opioids involved in almost two-thirds of fatalities (Rudd et al., 2016). In 2016, over 60,000 drug overdose fatalities were reported throughout the US (Centers for Disease Control and Prevention, 2017). New York City (NYC) experienced 1,374 drug overdose deaths in 2016 with 82% involving an opioid, with heroin being the most prevalent opioid involved (55%) and 44% involving fentanyl (New York City Department of Health and Mental Hygiene, 2017).

Reports suggest that public bathrooms,¹ such as those in fast-food restaurants, coffee shops, parks, and libraries, are among the most commonly used public injecting locations reported by people who inject drugs (PWID) in NYC (Injection Drug Users Health Alliance, 2015; New York City Department of Health and Mental Hygiene, 2010, 2013). Subsequently, numerous overdoses and overdose deaths have occurred in public bathrooms throughout NYC and the US (Associated Press, 2016a,b; Keilman, 2018; Raymond, 2017; VanGilder, 2017).

A previous study conducted by our team focused specifically on business managers’ encounters with drug use in NYC and revealed that many (58%) had encountered drug use in their business’s public bathroom (herein referred to as ‘business bathrooms’) within the past 6-months and that approximately a third (34%) of them had also encountered improperly disposed syringes (Wolfson-Stofko et al., 2017). Additionally, almost all (90%) had no training in overdose response and naloxone administration (ORNA), though a substantial minority (14%) had encountered someone who was unresponsive, presumably due to drug use, within the previous 6-months (Wolfson-Stofko et al., 2017). These findings suggest that business managers and staff (herein referred to collectively as a ‘service industry employee’ (SIE)) unwittingly become first-responders in the event of an opioid overdose and are not trained on how to respond to an overdose and administer naloxone. However, thirty-six states, including New York State, have amenable naloxone access and Good Samaritan laws that would permit ORNA training and naloxone distribution to SIEs (The Network for Public Health Law, 2017).

However, SIEs also face additional occupational risks associated with injection drug use occurring in their business bathrooms such as needlestick injuries and, in some cases, distressed PWID. Therefore, we designed an innovative intervention specifically for SIEs called Harm Reduction for Business Bathrooms (HRBB). This intervention included modules on 1) ORNA, 2) needlestick prevention and 3) strategies for de-escalating encounters with people who use drugs in their business bathrooms. Harm Reduction for Business Bathrooms was grounded in the Information-Motivation-Behavioral Skills (IMB) model (Fisher and Fisher, 1992; Fisher et al., 1996, 2009), where information, motivation, and behavioral skills interact synergistically to improve readiness to engage in the target behaviors. Specifically, HRBB was based on the assumptions that improving the behavioral skills necessary to prevent needlesticks and respond to opioid overdose, would thereby increase motivation and readiness to utilize ORNA skills in real-world settings. To our knowledge, no intervention has coupled needlestick prevention and de-escalation strategies with an ORNA intervention module specifically suited to the needs of SIEs who encounter drug use in their business bathrooms.

Additionally, no studies have evaluated the acceptability and evidence of efficacy (with respect to improving opioid overdose-related knowledge and attitudes, the intervention’s primary outcomes) of an ORNA intervention provided to SIEs whose businesses provide a bathroom for customers. This study sought to determine if the HRBB ORNA module was acceptable and efficacious in educating the target population on opioid overdose-related knowledge and improving opioid overdose management attitudes.

2. Methods

This study utilized a pre-/post-test concurrent nested mixed method trial design, and all activities were approved by the Institutional Review Board at National Development and Research Institutes, Inc. A Certificate of Confidentiality was obtained from the Department of Health and Human Services and written informed consent was obtained from each participant.

2.1. Participants

Inclusion criteria for participation included being 18 years or older and currently employed at a business located in NYC with a publicly accessible bathroom.

2.2. Description of the behavioral intervention

We used an Intervention Mapping process, a multi-step protocol for developing effective behavior change interventions, to develop the three-module intervention training curriculum (Bartholomew et al., 1998). As noted above, the IMB model was selected as the intervention’s guiding theoretical framework (Fisher and Fisher, 1992; Fisher et al., 1996, 2009).

The HRBB ORNA module was specifically designed to educate SIEs on how to respond to opioid overdoses and administer intranasal naloxone in accordance with training guidelines provided by the New York State Department of Health (New York State Department of Health, 2016) and Harm Reduction Coalition (Harm Reduction Coalition, 2012, 2016). This curriculum was constructed specifically for individuals with no medical training and minimal familiarity with ORNA. Sub-sections of the ORNA intervention module included information on 1) common prescription and street opioids (i.e., – Oxy-Contin, Vicodin, methadone, heroin, fentanyl, etc.), 2) signs of an opioid overdose (i.e., – unresponsive to verbal and physical stimuli, loss of consciousness, faint or no breathing, blue lips, blue fingertips, etc.), 3) New York State (NYS) naloxone access and Good Samaritan laws permitting naloxone possession and administration by laypeople, 4) how to assemble and administer intranasal naloxone and communicate with emergency services, 5) how to perform a sternum rub and place someone in the recovery position, and 6) the importance of why one should always respond as if fentanyl was involved (i.e., – rapidly, multiple doses, etc.)

Particular attention was placed on explaining New York State’s naloxone access and Good Samaritan laws that permitted naloxone distribution to SIEs and provided them, their colleagues and PWID with some legal protections against arrest, charges, and prosecution in regards to paraphernalia and drugs found at the scene of an overdose (The Network for Public Health Law, 2017). Additional recommendations were made on how to notify emergency services in order to prevent law enforcement involvement in a health emergency. New York State’s Good Samaritan law does not protect individuals who have outstanding warrants or those in violation of probation or parole. Accordingly, SIEs were instructed to tell emergency service operators that they found someone who is unresponsive and unconscious and not to mention that drugs may have been involved (in part because drug use may be assumed). Furthermore, reporting drug use may lead to a law enforcement response in addition to EMS which could escalate the situation, leading to the arrest of an overdose victim and/or preventing the individual from receiving medical attention. We wish to note that as a pilot study, the study was not designed to assess SIE’s interactions with PWID post-intervention, but instead focused on increasing information on and improved attitudes toward overdose, and increase readiness to intervene with PWID in future real-life encounters.

This ORNA module was presented using a PowerPoint presentation and interactive exercises demonstrating naloxone administration and completed in approximately 20 min. The intervention was conducted by the Principal Investigator (BWS) who is certified by NYS to train and distribute naloxone through an opioid overdose prevention program at one of the first author’s affiliated institutions (NDRI). This ORNA module has been made available by the authors in the Supplementary Materials section.

2.3. Procedures

Participants were recruited through convenience sampling and through contacts from previous studies focused on drug use occurring in business bathrooms (Wolfson-Stofko et al., 2017) in February 2017. As part of recruitment, researchers explained the intervention to an SIE who agreed to act as a liaison between researchers, business owners, and their fellow SIEs. If the business owners agreed to participate, the liaison sent an email to all employees on behalf of the research team inviting them to participate in the intervention. A consensus was then reached with SIEs and business owners on the most appropriate dates and times to conduct the training.

Eight different businesses (one café, one restaurant, two fast-food and four coffee shops) expressed interest, but only two received approval from business owners to participate. The two businesses that agreed to participate were both independent coffee shops, each with multiple locations in NYC. One business agreed to conduct both a morning and afternoon training on the same day in early March 2017 to accommodate SIEs who worked morning and afternoon shifts (eight participated in the morning session and four in the afternoon session). This business featured six different store locations consisting of nine managers and three staff (n = 12) that participated in the study.

The second business scheduled an afternoon training in early May 2017 and six employees (n = 6), four managers and two staff, representing three different store locations, participated in the training.

In total, 18 SIE (N = 18) participated amounting to 13 managers and 5 staff from nine different business locations.

The intervention was preceded by a pre-test structured assessment battery lasting approximately 15 min that included personal and employment demographics and inquired about their encounters with people using drugs in their business bathroom and their opioid overdose knowledge and attitudes.

Following the completion of the full intervention (~1.5 h for all 3 modules), all participants participated in an audio-recorded focus group (N = 3 focus groups) with their cohort (~30 min each) about their thoughts and perspectives on the training. After finishing the focus group, participants were given the opportunity to ask any remaining questions. Participants then completed a post-test assessment battery (~15 min) including an evaluation survey of the training.

All participants completed the pre- and post-test assessment batteries as well as the focus groups. They were compensated $40 for participation and were provided with their own take-home naloxone kit free of charge.

2.4. Measures

The Opioid Overdose Knowledge Scale (OOKS) and Opioid Overdose Attitudes Scale (OOAS) were developed to evaluate take-home naloxone training programs (Williams et al., 2013). Questions in regards to intramuscular naloxone were omitted because business employees were solely trained on the use of intranasal naloxone. The modified OOKS and OOAS were used to measure the primary outcomes of opioid overdose-related knowledge and attitudes, respectively, and correspond tightly with the constructs central to the IMB model. A modified version of the Client Satisfaction Survey was used to evaluate acceptability (Huba et al., 1997).

2.4.1. Opioid-overdose related knowledge

The modified OOKS consisted of a true/false questionnaire that was scored out of a maximum of 39-points from 39-items. These items used a ‘yes/no or don’t know’; or ‘true/false or don’t know’ response structure with correct answers receiving one-point and incorrect or ‘don’t know’ receiving no-point (Williams et al., 2013). The OOKS consists of four sub-scales that measure opioid overdose risks, signs, actions, and naloxone use. The OOKS has strong internal reliability (Cronbach’s α = 0.83), test-retest reliability (ICC = 0.90), and concurrent validity (r = 0.51, p < 0.001) (Williams et al., 2013).

2.4.2. Opioid-overdose related attitudes

The modified OOAS consisted of a 5-point Likert-type scale (completely disagree, disagree, unsure, agree, completely agree) scored 1–5, respectively. The questionnaire was scored out of a maximum of 115-points from 23-items (Williams et al., 2013). The OOAS consists of three subscales that measure an individual’s competence, concerns (about intervening), and readiness to intervene in the event of an opioid overdose. The OOAS has previously demonstrated internal reliability (Cronbach’s α = 0.90), test-retest reliability (ICC = 0.82) as well as concurrent validity (r = 0.51, p < 0.001) (Williams et al., 2013).

2.4.3. Acceptability

To assess acceptability, after the completion of the intervention, focus group, question and answer session, and post-test, participants engaged in a focus group (described below) and anonymously rated the intervention using a version of the Client Satisfaction Survey, Likert-type scale, modified specifically for this study (Huba et al., 1997). The rating scales ranged highest to lowest from: ‘excellent’, ‘very good’, ‘good’, ‘fair’, and ‘poor’ or ‘very helpful’, ‘helpful’, ‘not very helpful’, and ‘not helpful at all’. The survey also provided a section for anonymous free response. Both the Client Satisfaction Survey and focus group data were used to assess the acceptability of the intervention.

2.5. Focus group

Focus groups were conducted after the training but prior to the post-test assessment battery. Focus groups are ideal for obtaining community-wide perspectives and experiences (Stewart et al., 2006) and were utilized to further understand participants’ perspectives on the intervention and potential changes in attitudes towards responding to an opioid overdose in real-world encounters post-intervention. A semi-structured interview guide informed by similar studies (Marshall et al., 2017; Wagner et al., 2016, 2014) included questions such as, ‘How does being trained on how to recognize an overdose and administer naloxone make you feel?’; ‘What did you think about the information on the importance of remaining calm?’; ‘Do you think it would be worthwhile to train all of your staff and why?’; ‘What did you think of the hands-on naloxone training?’; “Do you believe that you could respond to an overdose and administer naloxone?”; and, ‘How can we improve this training?’

2.6. Quantitative data analysis

Descriptive statistics, paired t-tests, Wilcoxon signed rank tests, and effect size was computed using IBM SPSS Statistics 23 (ibm.com). Variables in regards to the participant’s encounters with drug use, unresponsive individuals, etc. in their business were derived from a previously conducted by our study team (Wolfson-Stofko et al., 2017). Pre-/post-test scores were compared to determine significance, confidence intervals, and effect size.

2.7. Qualitative data analysis

Interviews were digitally recorded, professionally transcribed, cleaned to remove vocalized pauses and filler utterances, and then entered into a software analysis program (MAXQDA, maxqda.com) for coding and analysis. The transcripts were analyzed using a hybrid deductive and inductive approach informed by the tenets of grounded theory (Charmaz, 2000; Glaser and Strauss, 1967; Strauss and Corbin, 1990) that aimed to develop a typology of perspectives described by SIE in regards to the intervention. Two authors (BWS and LE) with qualitative expertise separately coded one transcript, using a preliminary list of a priori codes based on topical domains addressed in the interview protocol (e.g., willingness to intervene, empowerment, etc.) adapted from similar studies (Marshall et al., 2017; Wagner et al., 2014). The coders met regularly during this process to discuss emergent themes and the addition of new code categories (e.g., importance of staying calm, requesting emergency medical services (EMS) and not police, etc.). A priori codes were augmented with a posteriori codes by consensus until a final code list was established. The final code list guided the coding and analysis of the remaining transcripts and the re-coding of the initial transcript, a task conducted by both BWS and LE.

3. Results

3.1. Demographic and background characteristics

A total of 18 participants (N = 18) enrolled in this study. Half identified as female with a mean age of 30 years (range = 22–38 years). Twelve identified as White, three as Asian, two as Latinx, and one as Black. Participants had worked in the service industry for an average of seven years (SD = 4.32 years, range = 1–10 years) and at their current location an average of two years (SD = 1.80 years, range = 1 month–5.5 years). Eleven had Bachelor’s degrees, three had Master’s degrees, one had an Associate’s degree, one had some college, one was still in college, and one had a high school diploma. Only one SIE had been previously trained in ORNA a few years before, and some were familiar with naloxone.

All business locations had single-stalled, unlocked public bathrooms except for one which had to be unlocked by staff. Twelve of the SIE stated that their business bathrooms had a deadbolt that could be unlocked with a master key.

Each participant was asked to independently recall and record the number of encounters they had with drug use occurring in their business bathroom as well as their encounters with paraphernalia and the actions taken in response to drug use throughout their tenure at these specific businesses. Participants reported that their businesses averaged three encounters with drug use per month (SD = 2.97 encounters per month, range = 0–12.5 encounters per month). A total of 29 encounters with unresponsive individuals occurred (range = 0–12 per store location), but no deaths related to drug use had occurred within the businesses. A total of 89 syringes were encountered in the past year (range = 0–36 per store location per year). Participants collectively made a total of 104 calls to emergency services in relation to drug use (72 calls made solely for law enforcement (range = 0–24 calls per year), 14 calls solely for EMS (range = 0–12 calls per year) and 18 calls requesting both law enforcement and EMS) per year.

3.2. Overdose response and naloxone administration component

3.2.1. Change in opioid overdose-related knowledge

Fig. 1 shows comparisons of pre-/post-test OOKS scores. Participants demonstrated an improvement in their total OOKS score after the intervention with a mean increase of 6.94-points (p < 0.01, Cohen’s d = 1.45) (Table 1). The most profound changes were in the subscales of naloxone use (p < 0.01, Cohen’s d = 1.89) with a mean increase of 4.44-points and signs of an overdose with a mean increase of 1.60-points (p < 0.01, Cohen’s d = 0.80).

Fig. 1.

Comparison of service industry employees’ pre-/post-test OOKS scores.

Table 1.

Comparison of SIEs’ pre-/post-test OOKS scores (n = 18).

	Pre-Test Mean (SD)	Post-Test Mean (SD)	p-value	Wilcoxon	95% CI	Cohen’s d
OOKS Totala	25.9 (4.30)	32.8 (3.33)	< 0.01*	< 0.01**	−9.32, −4.57	1.45
Risk Factorsb	6.39 (1.79)	6.61 (2.57)	0.71	0.75	−1.46, 1.01	0.09
Signsc	6.56 (1.89)	8.11 (1.23)	< 0.01*	< 0.01**	−2.52, −0.59	0.80
Actionsd	9.67 (0.97)	10.4 (0.70)	0.02*	0.02*	−1.28, −0.16	0.64
Naloxone Usee	2.28 (2.19)	7.72 (0.75)	< 0.01*	< 0.01**	−5.62, −3.27	1.89

^a Max Score = 39.

^b Max Score = 9.

^c Max Score = 10.

^d Max Score = 11.

^e Max Score = 9.

^* The mean difference is significant at the p = 0.05 level.

^** The mean difference is significant at the p < 0.01 level.

Participants achieved a small increase in knowledge on the two other subscales. Actions to be taken in an opioid overdose increased by 0.72-points (p = 0.02, Cohen’s d = 0.64) and opioid overdose risk factors by 0.22-points (p = 0.71, Cohen’s d = 0.09). As a sensitivity analysis, we considered a Wilcoxon signed rank test for all OOKS scales and subscales, and the inferences had minor variations.

The qualitative data below further elaborated on SIE’s opioid overdose-related knowledge.

3.2.2. Recognizing signs of an overdose and administering naloxone

Participants were asked to reflect on the usefulness of the information provided by the ORNA module. One participant elaborated on her perception of the module and training methods:

“Really helpful. I liked that it involved a lot of different methods of trainings such as a presentation…and then a demonstration…It’s really helpful to do that. I like it that we’re left with something to use (naloxone kit). We’ve been taught how to use it. I thought it was very clear. I feel definitely more confident now than I was in the beginning of the class about helping someone in an overdose situation for opioids.”

[Skandar, female]

Other participants stressed the importance of the hands-on training of how to assemble the intranasal naloxone that was provided:

“The way that I would figure out these instructions on this bag [naloxone kit carrying case], someone could die. So, the hands-on instruction is super good.”

[Rayna, female]

“Once you explain how to use it [intranasal naloxone], it’s pretty easy for anyone to be able to do.”

[Lucy, female]

One participant further suggested that her business should have at least one ORNA trained person on staff at each location, similar to how NYC requires her business to have at least one person at each business location during operating hours certified in safe food preparation (New York City Department of Health and Mental Hygiene, 2018):

“I think ideally we would have someone on staff trained how to use it [naloxone], the same way that we have our food handlers card person, be like long-term goals.”

[Lucy, female]

Lucy’s long-term goal of having at least one ORNA trained SIE at all locations during operating hours suggests that she views the ORNA intervention as a necessary service that she believes would benefit her business and its patrons. These perspectives suggest the information was positively received.

3.2.3. Good samaritan law and motivations for contacting emergency medical services

As noted above, prior to discussing how to contact emergency services, the intervention discussed New York State’s naloxone access and Good Samaritan laws and how they applied to SIEs and overdose victims when responding to an overdose. Participants reflected on the issues that could arise from law enforcement involvement:

“I’m pretty skeptical about a lot of things [that law enforcement does], and then the person…I’m trying to help is probably 50 times more skeptical of them [law enforcement] than me. I don’t want to get their kids taken away. I don’t know what their situation is. I think that’s the least comfortable part of everything.”

[Poppy, female]

“Well, in the part where we’re calling [911] – trying to just get the ambulance, the cops might just make [the response] worse or more chaotic.”

[Prox, male]

These comments suggest that some of the SIEs were empathetic to the plights of the people that use drugs in their business bathrooms and the effects that a drug possession or other charge could have on a PWID’s well-being. Though some felt that the Good Samaritan protections did not go far enough in protecting individuals who overdose, SIEs concluded that the need for medical services supersedes their concerns about potential legal repercussions for the naloxone recipient.

3.2.4. Importance of staying calm

Information on the importance of remaining calm when responding to an overdose was a key aspect that was stressed during the intervention. The trainer stressed the idea of ‘grounding’ oneself in the event of an emergency, such as by taking a deep breath prior to responding to an unresponsive individual, so that one can collect their thoughts and focus on the situation at hand. Talley further elaborated on the importance of teaching ‘grounding’ techniques:

“The way that you [the trainer] stressed staying calm, I think…[people] kind of forget that’s a choice that you can actually make when you’re stressing out…[and] actually take control of that.”

[Talley, female]

Rayna also stated that it would be easier for her to remain calm when responding to an overdose because the training informed her that naloxone would not harm a person whether or not they are having an overdose:

“And it’s scary when you’re talking about medical stuff and stuff where someone could die. But I think all the normalizing of it… like no way [the person overdosing] is going to get hurt in this process makes it so much easier to stay calm.”

[Rayna, female]

3.2.5. Change inopioid overdose-related attitudes

Fig. 2 shows comparisons of pre-/post-test OOAS scores. All participants demonstrated an improvement in their total OOAS score after the intervention. On average, total scores increased by 22.9-points (p < 0.01, Cohen’s d = 2.45) after the intervention (Table 2).

Fig. 2.

Comparison of service industry employees’ pre-/post-test OOAS scores.

Table 2.

Comparison of SIEs’ pre-/post-test OOAS scores (n = 18).

	Pre-Test Mean (SD)	Post-Test Mean (SD)	p-value	Wilcoxon	95% CI	Cohen’s d
OOAS Totala	74.1 (7.50)	96.9 (6.42)	< 0.01*	< 0.01**	−27.5, −18.2	2.45
Competenceb	18.6 (5.79)	33.3 (3.31)	< 0.01*	< 0.01**	−17.6, −11.8	2.54
Concernsc	14.3 (2.30)	19.1 (2.85)	< 0.01*	< 0.01**	−6.42, −3.25	1.52
Readinessd	41.2 (4.41)	44.5 (3.28)	< 0.01*	< 0.01**	−4.80, −1.87	1.13

^a Max Score = 115.

^b Max Score = 45.

^c Max Score = 25.

^d Max Score = 50.

^* The mean difference is significant at the p = 0.05 level.

^** The mean difference is significant at the p < 0.01 level.

Most of this change can be attributed to the increase in participants’ competencies to manage an opioid overdose (14.7-points, p < 0.01, Cohen’s d = 2.54). There was also a positive change associated with participants’ readiness to intervene in an opioid overdose (p < 0.01, Cohen’s d = 1.13) by 3.33-points as well as a reduction (positive change) of participant’s concerns about managing an overdose (p <0.01, Cohen’s d = 1.52) by 4.83-points. As a sensitivity analysis, we considered a Wilcoxon signed rank test for all OOAS scales and sub-scales, and the inferences had minor variations.

The qualitative data below further elaborate on these changes in opioid overdose-related attitudes.

3.2.6. Motivation to intervene

No participants stated that they felt unable to intervene while many explicitly stated that they are willing to intervene in an overdose. One participant suggested that the knowledge and skills acquired in this intervention relieved the sense of helplessness that they’d experience while waiting for emergency services to respond during previous incidents:

“To know that in that time [after calling for emergency services], there is something that I actually can do to potentially save this person’s life – I feel so glad that I can do that. And so definitely, I will do that.”

[Alaska, male]

Though participants report being confident in their understanding of ORNA, some confided that they would be nervous when responding to an overdose:

“I’m not nervous about actually doing it. It’s just remembering how to do it. I know how to do it. It seems a lot less scary now.”

[Prox, male]

“I definitely feel better prepared. But I think I’d still be nervous that I’d fuck it up.”

[Latrice, female]

There was no indication that the participants would have been less nervous as a result of more training.

3.2.7. Empowerment

Many SIEs stated that this intervention made them feel “empowered” as similarly reported in previous studies (Marshall et al., 2017; Wagner et al., 2014). Participants also stated that being supplied with naloxone made them feel “equipped” so that they could act in the event of an overdose:

“Feeling confident and equipped to be an agent in saving someone’s life and that to a person is just invaluable…to be at work and – or somewhere else and know that you can do something.”

[Sevn, male]

“I learned a lot. It’s also a lot less scary than I think I thought it would be administering this. I feel like if a situation were it ever to come up, I would feel more empowered and confident about actually helping.”

[Alaska, male]

This sense of empowerment aligns with the improvement of quantitative measures in regards to competencies in managing an opioid overdose and having fewer concerns about responding to an opioid overdose.

3.3. Participant evaluation of acceptability of and other aspects of the intervention

Thirteen of the participants rated the overall intervention as ‘excellent’, three as ‘very good’ and two as ‘good.’ Most pertinent for this manuscript are the ratings pertaining to the ORNA module. Fifteen participants rated the information that they received on opioids as ‘very helpful’ and three as ‘helpful.’ The information provided on how to recognize an overdose was rated the same. Training on how to reverse an overdose with naloxone received 17 ratings of ‘very helpful’ (highest possible rating) and one as ‘helpful.’ Though the vast majority of participants rated the ORNA module of the intervention as highly favorably, some SIE made suggestions on how the intervention could be improved.

3.3.1. Reflections on the training

Participants were supportive of the design and topics covered in the intervention and supportive of its dissemination:

“I wouldn’t make any changes. I believe this should be taught to all employees.”

[Anonymous free response]

“I would like to see training become mandatory for businesses especially in NYC where we have a large overdose problem.”

[Anonymous free response]

Many participants also believed that public employees, particularly librarians, would benefit from this intervention being that they also manage public bathrooms where injection drug use occurs (Associated Press, 2016a; Bautista, 2015).

3.3.2. How to improve the intervention

The most common critique was limited opportunities to practice via role-playing or on a CPR manikin. Participants were given the opportunity to handle an intranasal injector and to do a test spray with water (instead of naloxone), but some feel that more practice would be beneficial. Another participant in one of the focus groups suggested that SIEs might benefit from hearing about the experiences of individuals who have had naloxone administered to them:

“We could benefit from hearing from someone in recovery or someone who has been through it before.”

[Latrice, female]

Other participants thought that a video of someone who had overdosed, been administered naloxone and experienced the withdrawal could provide could help quell fears about naloxone administration and in turn motivate ORNA.

4. Discussion

In the current research, we explored the evidence of efficacy and acceptability of an ORNA intervention module embedded in a larger three-module intervention program designed for SIEs. Participants found the intervention module acceptable, and many believed that it would be beneficial for other SIEs and their colleagues to be trained in ORNA as a part of company policy. Results from this study provide further evidence supporting the expansion of ORNA to a wider variety of individuals and occupations who regularly encounter overdoses. Participants in this study specifically suggested that other professionals who manage public bathrooms, such as librarians, would benefit from this intervention as well. This is of increasing significance as more reports emerge reaffirming the use of public bathrooms for injecting and subsequent overdoses (Associated Press, 2016a; Bautista, 2015; Bebinger et al., 2017; Seelye, 2016; Tracy, 2017).

This intervention was designed to be applicable to SIEs in US states that have enacted laws that protect lay people who possess and administer naloxone from criminal and civil liability, which at the time of writing includes 36 states (The Network for Public Health Law, 2017). This intervention might be best implemented at local or general staff meetings where many SIEs can be trained at once and be compensated for their time.

There are also public policy considerations that could help support the expansion of ORNA to businesses with public bathrooms. One participant suggested that ORNA be required for businesses similar to that of food protection training. New York City, for example, requires at least one person holding a Food Protection Certificate to be present in the business during all hours of operation to ensure food safety (New York City Department of Health and Mental Hygiene, 2018). A similar ORNA requirement would help ensure that all businesses with public bathrooms would have naloxone and at least one employee capable of responding to an opioid overdose during operating hours. Furthermore, public knowledge of businesses possessing naloxone would also allow employees and trained members of the public to respond to an opioid overdose occurring within the immediate vicinity outside of a business. Arguably, a standardized distribution system of naloxone, similar to that of automated external defibrillators (AED) in NYS (Centers for Disease Control and Prevention, 2012; State of New York, 1998), may prevent fatal overdoses within communities if residents know where they can obtain naloxone in the event of an opioid overdose. Due to the extent of the overdose epidemic in the US, such policy considerations are worth further exploration.

Furthermore, it is important to train SIEs in ORNA because they are unable to know how long a person has been unresponsive in their business bathroom or how long it will take for EMS to arrive which justifies the need for instantaneous action. This intervention instructs participants to call EMS immediately upon identifying a potential overdose, prior to naloxone administration, to ensure that EMS arrive as soon as possible. The average EMS response time for a life-threatening medical emergency in NYC in 2017 was 7.29 min (NYC 911 Reporting, 2018), similar to the US urban average of 7 min and a suburban average of 7.7 min while the rural average was 14.5 min (Mell et al., 2017). However, brain damage can occur from 3 to 5 min after breathing cessation (Harm Reduction Coalition, 2012) making SIE response and naloxone administration crucial and potentially lifesaving.

Some have argued that the naloxone recipient could become hostile after naloxone administration and endanger SIEs though there is minimal evidence suggesting that naloxone administration leads to physical violence though it can provoke discomfort and agitation in opioid-dependent individuals. Contacting EMS prior to naloxone administration helps to ensure that EMS arrive soon after an individual regains consciousness and provide a smaller window of time for an individual to flee the scene prior to EMS arrival.

It is also important to note that the training of SIEs and other professionals in ORNA is a response to public bathroom drug use and not a preventative measure. Previous research suggests that the operation of supervised injection facilities (SIF) could reduce such public injection (Hunt et al., 2007; Kral et al., 2010; van Beek and Gilmour, 2000; Wood et al., 2004) and are worthy of further research and implementation.

Future research on the HRBB intervention will include the evaluation of the needlestick prevention module as well as the expansion of the intervention to a larger sample of SIEs from a broader set of businesses (fast-food, laundromat, etc.) and SIEs in suburban and rural communities.

4.1. Limitations

The results from this study cannot be generalized for all of NYC being that this study used convenience sampling and contacts from previous studies and consisted of a small sample size. However, the goal of this study was to generate preliminary indications of efficacy and SIEs’ perspectives towards an ORNA intervention module designed specifically for SIEs. The results of this study may be biased; participants willing to take the time to participate in the intervention may be more receptive to public health interventions. Recall in regards to past encounters with drug use, paraphernalia, unresponsive individuals, law enforcement and EMS may vary depending on SIEs experiences and perceptions of substance use. A greater effort can also be made to specifically teach the terminology utilized by the scales to ensure that participants fully understand all terminology. There is also the possibility that employees provided socially desirable answers though researchers tried to offset this by trying to build rapport with participants throughout the intervention and by providing them with the opportunity to provide anonymous feedback. As noted above, the assessment of SIE’s actual response behaviors to PWID or naloxone use subsequent to training was beyond the scope of the study. However, the IMB model highlights how motivation to perform a new behavior (such as ORNA) is in large measure a function of sufficient knowledge, positive attitudes, and behavioral skills, including access to resources such as naloxone kits. For the purpose of the present study, we speculate that these factors will act synergistically to increase the chances that SIEs will respond appropriately to overdose events in their places of business. Indeed, study results support previous research suggesting that knowledge and attitudes towards ORNA can be improved by providing information, motivation, and behavioral skills to non-medical professionals’ (Behar et al., 2015; Green et al., 2008; Jones et al., 2014; Pade et al., 2016; Wagner et al., 2016; Williams et al., 2014). We will assess the effects of the training on behavioral outcomes over a longer period in future research.

5. Conclusion

This small study is the first to suggest evidence of efficacy in the training of SIEs in ORNA and that this module of the SIE-specific intervention is acceptable. The results from both the quantitative and qualitative components further support previous research that suggests that non-medical professionals are capable of identifying and managing opioid overdoses after a brief training.

Abbreviations:

AED:

automated external defibrillators

CPR:

cardio-pulmonary resuscitation

EMS:

emergency medical services

HRBB:

harm reduction for business bathrooms

IMB:

informational-motivational-behavioral skills model

OOAS:

opioid overdose attitudes scale

OOKS:

opioid overdose knowledge scale

ORNA:

overdose recognition and naloxone administration

PEP:

post exposure prophylaxis

PWID:

people who inject drugs

SIE:

service industry employee