- Email: [email protected]
Witnessed overdoses and naloxone use among visitors to Rikers Island jails trained in overdose rescue
Addictive Behaviors xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Addictive Behaviors journal homepage: www.elsevier.com/locate/addictbeh
Witnessed overdoses and naloxone use among visitors to Rikers Island jails trained in overdose rescue ⁎
Zina Huxley-Reichera, ,1, Lara Maldjiana, Emily Winkelsteina, Anne Sieglerb, Denise Paonea, Ellenie Tuazona, Michelle L. Nolana, Alison Jordanb, Ross MacDonaldb, Hillary V. Kuninsa a b
New York City Department of Health and Mental Hygiene, 42-09 28th Street, 19th Floor, Long Island City, NY 11101, United States New York City Health + Hospitals, 55 Water Street, 18th floor, New York, NY 10014, United States
H I G H L I G H T S prospective study of NYC jail visitors to Rikers Island trained in naloxone. • AOf6-month 283 participants enrolled, 14% witnessed at least one overdose. • Of the 283 participants enrolled, 10% administered naloxone at least once. • Thethenaloxone use is comparable to similar interventions for high-risk populations. • Training jail visitors is effective at reaching a population at risk of overdose. •
A R T I C L E I N F O
A B S T R A C T
Keywords: Naloxone Jail Overdose Rescue Criminal justice Opioid
With the opioid overdose mortality rates rising nationally, The New York City Department of Health and Mental Hygiene (NYC DOHMH) has worked to expand overdose rescue training (ORT) and naloxone distribution. This study sought to determine rates of overdose witnessing and naloxone use among overdose rescue-trained visitors to the NYC jails on Rikers Island. We conducted a six-month prospective study of visitors to NYC jails on Rikers Island who received ORT. We collected baseline characteristics of study participants, characteristics of overdose events, and responses to witnessed overdose events, including whether the victim was the incarcerated individual the participant was visiting on the day of training. Bivariate analyses compared baseline characteristics of participants who witnessed overdoses to those who did not, and of participants who used naloxone to those who did not. Overall, we enrolled 283 participants visiting NYC's Rikers Island jails into the study. Six months after enrollment, we reached 226 participants for follow-up by phone. 40 participants witnessed 70 overdose events, and 28 participants reported using naloxone. Of the 70 overdose events, three victims were the incarcerated individuals visited on the day of training; nine additional victims were recently released from jail and/or prison. Visitors to persons incarcerated at Rikers Island witness overdose events and are able to perform overdose rescues with naloxone. This intervention reaches a population that includes not only those recently released, but also other people who experienced overdose.
1. Introduction Mortality rates from unintentional opioid overdose continue to increase across the United States. Between 2000 and 2014, the rate of opioid overdose deaths tripled (Wheeler, Jones, Gilbert, & Davidson, 2015). Opioid overdose deaths are preventable with a multi-pronged approach, including naloxone distribution to laypeople. Naloxone is an opioid antagonist medication that reverses the effect of opioid intoxication (overdose) and restores breathing. In New York State, opioid ⁎
1
overdose prevention programs, like overdose prevention programs in other jurisdictions, educate laypeople to recognize and respond to an overdose, including how to administer naloxone. Distribution of naloxone to laypeople at risk of overdose is cost-effective and reduces overdose mortality (Coffin & Sullivan, 2013; Gaston, Best, Manning, & Day, 2009; Sherman et al., 2008; Strang et al., 2008; Wagner et al., 2010; Walley et al., 2013). In the United States, there have been over 26,000 reversals reported and over 150,000 kits dispensed since 1996 (Wheeler et al., 2015).
Corresponding author. E-mail address: [email protected] (Z. Huxley-Reicher). Present address: Icahn School of Medicine at Mount Sinai; 1 Gustave L Levy Place, New York, NY 10029, United States.
https://doi.org/10.1016/j.addbeh.2017.11.029 Received 25 July 2017; Received in revised form 6 November 2017; Accepted 15 November 2017 0306-4603/ © 2017 Published by Elsevier Ltd.
Please cite this article as: Huxley-Reicher, Z., Addictive Behaviors (2017), https://doi.org/10.1016/j.addbeh.2017.11.029
Addictive Behaviors xxx (xxxx) xxx–xxx
Z. Huxley-Reicher et al.
verbal consent which included education about the study's risks and benefits as well as confidentiality and study procedures. All trained individuals received naloxone kits as they exited Rikers Island Central Visit Center, and after completing their visit to the incarcerated person. Study participants included all individuals who 1) participated in ORT, 2) gave verbal consent to participate in the study, 3) provided demographic (age, race/ethnicity and gender) and contact information in either Spanish or English, and 4) received naloxone after their visit. IRBs at the NYC DOHMH and NYC Health and Hospitals approved the study.
In order for naloxone distribution to reduce overdose mortality effectively, distribution to populations at highest risk of overdose should be prioritized. In many jurisdictions, including New York City (NYC), syringe exchange programs (SEPs) accomplish this goal by engaging and training people who actively use drugs to recognize and respond to an overdose, including how to administer naloxone. A meta-analysis of community naloxone trainings among people who use drugs reported naloxone use rates that ranged from 5 to 13% in the three months after naloxone receipt (McCauley, Aucott, & Matheson, 2015). NYC SEP participants trained to use naloxone reported higher rates; 25% used naloxone to reverse an overdose within 12 months of training (Siegler et al., 2017). Following release from carceral settings, individuals are at risk of fatal overdose due to reduced opioid tolerance after a period of abstinence. In NYC, people with a recent history of incarceration are eight times more likely to die from drug-related causes in the two weeks after incarceration than non-incarcerated individuals (Lim et al., 2012). Similar findings are reported in other parts of the United States and internationally (Binswanger et al., 2007; Farrell & Marsden, 2008; Bird and Hutchinson, 2003; Stewart, Henderson, Hobbs, Ridout, & Knuiman, 2004; Merrall et al., 2010; Winter et al., 2015). Additionally, of the > 200,000 individuals released from NYC jails between 2011 and 2016, 36,000 (18%) had an opioid use disorder (unpublished data, Correctional Health Service, NYC Health and Hospitals). In response to the high risk of overdose following incarceration, a number of programs dispense naloxone to formerly incarcerated persons upon re-entry. One program in Scotland found reductions in overdose mortality rates following the implementation of naloxone dispensing to people at prison discharge (Bird, Parmar, & Strang, 2015). In the U.S., programs in Rhode Island, California, Washington, New Mexico and New York State have begun to provide naloxone to releases (Beletsky et al., 2015). In the NYC jail system, like other correctional systems, a number of logistical barriers have thus far precluded naloxone distribution to incarcerated people upon release. By training visitors, we sought to prevent fatal overdose among persons recently released from jail on Rikers Island, where approximately three-quarters of NYC's jails are located. Beginning in 2014, the New York City Health Department began a program to offer training to visitors in overdose recognition and naloxone administration prior to their jail visit. Those trained received a naloxone kit after their visit, as they exited the Rikers Island Central Visit Center. To determine whether naloxone distribution to jail visitors would lead to overdose rescues among recently incarcerated people, we conducted a prospective study of: 1) the prevalence of witnessing overdose among jail visitors in the 6 months following the visit; 2) naloxone use rates by jail visitors; and 3) the proportion of naloxone administrations by jail visitors that were to individuals recently incarcerated at Rikers Island.
2.2. Overdose rescue training and naloxone dispensing NYC Health Department employees or syringe-exchange outreach workers conducted ORT. ORT lasted from two to five minutes and included instruction on signs of opioid overdose, steps for responding to opioid overdose how to administer intranasal naloxone, and protections provided by the 911 Good Samaritan Law—a New York State law that protects individuals from prosecution when calling 911 in case of an overdose (Drug Policy Alliance, 2015). If time permitted, trainers also instructed participants in rescue breathing and overdose prevention tips. Due to restrictions on property permitted inside individual jail facilities, staff gave naloxone kits to trained visitors after their visit, as they exited the jail complex, often three to four hours after initial training. Each naloxone kit contained two doses of 1 mg/mL naloxone with two nasal adaptors, disposable gloves, an alcohol swab, instructions for use of the product, and a rescue breathing mask. 2.3. Follow up procedures We asked study participants to provide at least one mechanism for follow up contact (phone number or email address) and a mailing address to receive incentives. Participants were reached by phone call or email to confirm contact information at one month and three months after enrollment. At six months, we administered a closed-ended survey by phone. Participants who agreed to be contacted via text message also received texts each month to maintain contact. At one, three, and six months, we attempted to reach participants by phone-call up to three times, and by email and post card if unsuccessful by phone. 2.4. Incentives Following enrollment at the Central Visit Center, we gave participants round-trip subway cards valued at $5.50. Participants received an additional subway card upon completion of the one-month check-in, two subway cards upon completion of the three month check in, and a $20 gift card upon completion of the six-month follow-up survey. The maximum incentive was $42. 2.5. Measures
2. Methods We collected demographic information immediately following ORT. Demographic characteristics of study participants included age, gender, race/ethnicity, and borough of residence. The questions on the sixmonth survey included questions about witnessing overdose, use of naloxone, incarceration history of overdose victim, and relationship to overdose victim.
2.1. Sample and study enrollment We conducted a prospective observational study of individuals visiting incarcerated persons at Rikers Island who completed overdose rescue training (ORT) and received naloxone. Participants were recruited by convenience sampling from the Rikers Island Central Visit Center during five consecutive days in August 2015. These represented all days available for visiting in a week (not every day is open and days are divided by last name of the individual being visited). Upon entry into the Rikers Island Central Visit Center, program staff offered all visitors ORT, regardless of the sentencing status of the incarcerated person they were visiting, while they awaited transport to the individual jails in the complex. Staff invited individuals who agreed to be trained to participate in the study. Staff then obtained informed
2.6. Outcomes The three outcomes of interest were: (1) witnessing a drug overdose, (2) administering naloxone, and (3) whether the overdose victims were the individuals being visited at Rikers on the day of training. 2.6.1. Witnessing drug overdose At six months, participants reported whether they had witnessed 2
Addictive Behaviors xxx (xxxx) xxx–xxx
Z. Huxley-Reicher et al.
Fig. 1. Flow chart of participants in Rikers Island visitor overdose rescue training.
ORT and naloxone administration during any witnessed overdose event dichotomously. To calculate the rate of witnessed overdoses, the number of participants who witnessed an overdose in the six-month evaluation period was divided by the total number of study participants. We calculated the rate of naloxone administrations by dividing the number of participants who used naloxone in a six-month time period by the total number of study participants.
any overdoses since study enrollment. An overdose was defined to study participants as “unresponsive or unable to be woken up, collapsing, having blue skin color, having difficulty breathing, losing consciousness or dying while using drugs.” For each witnessed overdose, participants reported (1) their relationship to victim, (2) survival status of victim, (3) recent incarceration status of victim (whether they had been in jail or prison in the last year), (4) and their responses to the event. 2.6.2. Naloxone administration Participants reported naloxone administration for each witnessed overdose. For overdoses where a layperson administered naloxone, participants indicated whether they or another bystander administered naloxone. If an overdose was witnessed, but naloxone was not administered, we asked the main reason it was not administered. Participants who did administer naloxone reported whether they observed a response in the victim after the naloxone administration. A secondary outcome assessed whether participants or another bystander called 911. We also asked participants if they obtained a replacement naloxone kit following administration.
2.7.2. Event-level variables We categorized whether the overdose victim was the person visited at the time of ORT training dichotomously. We categorized past-year incarceration of the overdose victim trichotomously as yes, no or don't know. Relationship to overdose victim included four categories: friend, family member, acquaintance and stranger. We treated survival status as assessed by participants and whether 911-calling as dichotomous variables. Descriptive analyses of study participants and overdose events were conducted. Using univariate logistic regression, we calculated 95% confidence intervals and p-values to determine which variables (age, gender, race/ethnicity, day of enrollment, and borough of residence) were associated with witnessing overdose and administering naloxone. The small number of participants who reported witnessing an overdose where the victim was the individual visited on the day of training precluded bivariable analysis. Statistical significance was defined as p < 0.05; analyses were performed using SAS 9.2 (Cary, North Carolina).
2.6.3. Overdose victim's incarceration history For each witnessed overdose event, participants reported whether the victim was the individual they were visiting on the day they were trained at Rikers. If not, research staff asked whether the victim had been recently released (within the last year) from jail or prison. 2.7. Data analysis We compared demographic differences between those who completed and did not complete the six-month survey using a chi-square to determine whether loss to follow-up could have biased study findings. The analyses included both individual (by participant) and the event (by overdose event) levels.
3. Results During the five days of enrollment, 382 visitors were trained in overdose rescue at the Rikers Island Visit Center. A total of 335 (88%) returned to request a naloxone kit after their visit, and 283 (85%) consented to participate in the study. A total of 226 participants who consented to participate (80%) were reached and completed the sixmonth follow up (Fig. 1). Our study population was 83.2% female, 43.6% African American, 38.2% Hispanic and the mean age was 38.4 years old. We reached younger visitors, aged 16–24 years, less frequently to complete the 6-month follow-up survey compared to the sample as a whole (69% vs 80%, p < 0.04).
2.7.1. Individual-level variables We collapsed age (years) into five categories, 16–24, 25–34, 35–44, 45–54 and 55 and older. We categorized race/ethnicity as Hispanic/ Latino, non-Hispanic Black/African American, non-Hispanic White, and other. We categorized enrollment date as weekday or weekend. We categorized witnessing at least one overdose in the six months following 3
Addictive Behaviors xxx (xxxx) xxx–xxx
Z. Huxley-Reicher et al.
Table 1 Characteristics of participants who completed overdose rescue training (ORT) and completed the six-month survey.
Study population (n = 283) Age (mean) 16–24 25–34 35–44 45–54 55 + Gendera Male Female Race/ethnicity Hispanic/Latino Non-Hispanic White Non-Hispanic Black/African American Otherc Borough of residenced Bronx Brooklyn Manhattan Queens Staten Island Other Day of overdose rescue training Weekday Weekend Mean number of witnessed overdoses (range)
Total sample who completed 6 month survey
Witnessed an overdosea
Naloxone administeredb
n (%e)
n (%f)
n (%g)
226 (79.9%)
40 (14.1)
28 (9.9)
38.4 40 (17.7) 58 (25.7) 52 (23.0) 40 (17.7) 36 (15.9)
40.9 5 (12.5) 8 (20.0) 11 (27.5) 6 (15.0) 10 (25.0)
40.8 4 (14.3) 5 (17.9) 9 (32.1) 2 (7.1) 8 (28.6)
38 (16.8) 188 (83.2)
7 (17.5) 33 (82.5)
5 (17.9) 23 (82.1)
86 (38.2) 13 (5.8)
17 (43.6) 3 (7.7)
13 (48.2) 1 (3.7)
98 (43.6)
18 (46.2)
13 (48.2)
28 (12.4)
1 (2.6)
0
60 55 31 37 21 22
12 (30.0) 10 (25.0) 6 (15.0) 4 (10.0) 4 (10.0) 4 (10.0)
11 (39.3) 7 (25.0) 5 (17.9) 0 2 (7.1) 3 (10.7)
30 (75.0) 10 (25.0) 1.8(1–4)
20 (71.4) 8 (29.6) 1.9 (1-4h)
(26.6) (24.3) (13.7) (16.4) (9.3) (9.7)
170 (75.2) 56 (24.8) 0.3 (0–4)
Table 2 Characteristics of overdose events. Event characteristics
Number of events (%)
Total events Victim characteristics Recently released from jail or prisona Yes No Don't Know Was visited by study participant on day of naloxone receipt Relationship Familyb Friend Stranger Acquaintancec Survival Yes No Do not know Responses 911 called by anyone Naloxone given by participant by another layperson
70 (100)
12 (17.1) 25 (35.7) 33 (47.1) 3 (4.3)
12 14 32 12
(17.1) (20) (45.7) (17.1)
61 (87.1) 4 (5.7%) 5 (7.1%) 56 (80) 50 (71.4) 5 (7.1)
a Recently is defined as within the last year (n = 6 events were < 6 months after release, n = 4 events were > 6 months after release and n = 2 were unknown). b Family includes: victim's parent, sibling, relative, and spouse/partner. c Acquaintance includes: client, neighbor and acquaintance.
relevant percentages but not the odds ratios or p values).
3.2. Naloxone administration During the 70 witnessed overdose events, a lay person administered naloxone in 55 events (79%); study participants, specifically, administered naloxone in 71% of events. Overall, 28 (10%) study participants reported administering naloxone at least once during the study period. Either a participant or another bystander called 911 in 56 (80%) of the witnessed overdose events (Table 2). The main reasons that participants reported for not administering naloxone at overdose events (n = 20, 29%) were: emergency personnel arrived first (n = 8; 40%), the participant did not have naloxone on hand (n = 5; 25%), and someone else administered naloxone (n = 3; 15%) (Table 3). None of the measured participant characteristics were associated with administering naloxone (see Table 1). Of the 55 overdoses where lay people administered naloxone, participants reported that the victim responded to the naloxone (“woke up after they were given naloxone”) in 50 events (91%). Of the four events where the victim did not respond to naloxone administration, the victim did not survive in three, and had unknown survival in one. In one event, the participant did not answer the question about victim's response to naloxone administration.
a
No variables were significantly associated with witnessing an overdose. No variables were significantly associated with using naloxone. c Includes those who reported Asian, Native American, Mixed Race. d Residence at the time of recruitment. e Column %- denominator is the total n = 226 for those who completed the 6 month survey. f Column % - denominator is the total n = 40 for those who have witness an OD. g Column % - denominator is the total n = 28 of those who administered Naloxone. h All individuals who witnessed 3 (n = 6) and 4 (n = 3) ODs used naloxone in at least one of the witnessed OD events. b
3.1. Witnessing drug overdose Of the 226 participants reached at six months, 40 (14% of the total enrolled n = 283) had witnessed at least one overdose during the study period (Table 1). Twenty-three (23, 56%) participants witnessed one overdose, nine (23%) witnessed two, five (13%) witnessed three and three (8%) participants witnessed four events during the six months following ORT. In total, study participants witnessed 70 overdose events. Survival data was available for 65 of the 70 events (92%) (Table 2); of those 65 events, the victim survived in 61. The relationship between the participant and victim varied: 46% of the victims were strangers to the participant, 20% were friends, 17% were family members, and 17% were acquaintances. In 47% of events the victim's incarceration history was unknown; in 17% of events, victims were recently released from jail or prison. In three (4%) of the events, the overdose victim was the person visited on the day the participant received ORT. No variables measured (age, gender, race/ethnicity, borough of residence, or weekday/weekend enrollment) were associated with witnessing an overdose (see Table 1; to conserve space, we show the
Table 3 Main reason naloxone was not administered by participant at witnessed OD events. Reason
n (%)
Total events Emergency personnel arrived first Didn't have naloxone on hand Someone else administered naloxonea Other Had already used kit
20 (100) 8 (40) 5 (25) 3 (15) 3 (15) 1 (5)
a In an additional 2 events naloxone was administered by another bystander; but this was not the main reason the participant didn't administer.
4
Addictive Behaviors xxx (xxxx) xxx–xxx
Z. Huxley-Reicher et al.
4. Discussion
naloxone refills after receipt of brief overdose rescue training (Behar, Santos, Wheeler, Rowe, & Coffin, 2015). This study had additional limitations. Due to the restrictions on what can be brought into jail facilities, we distributed naloxone hours after the initial training and 12% of those trained did not receive a kit. In order to minimize participants' burden, we limited the number of questions on the six-month survey. This resulted in limited data about the witnessed events and the use of naloxone. Also, we found no associations of independent variables with our outcomes (witnessing overdose and using naloxone). While this may demonstrate a real pattern, it is likely due to the fact that our study was not powered sufficiently to detect differences. In addition, 20% of the initial study population were lost to follow up; it is possible that participants who responded at six-months were more likely to witness and respond to overdose, thereby biasing our results upwards. By including the rate among the total study population, rather than the responders only, we sought to minimize this bias.
Visitors to the Rikers Island Central Visit Center who received ORT had a high frequency of witnessing overdoses post-training. Although this intervention successfully reached a population that witnessed overdoses in rates similar to other overdose rescue training programs, few of the overdoses witnessed were among people visited on the day of training. Study participants witnessed overdoses at nearly the same rate reported in previous studies of naloxone training programs (Bennett & Halloway, 2012; McAuley, Lindsay, Woods, & Louttit, 2010; Strang et al., 2008; Tobin, Sherman, Beilenson, Welsh, & Latkin, 2009). Our finding that visitors of individuals incarcerated at Rikers Island who are trained in ORT witness and respond to overdoses at high rates demonstrates that distributing naloxone to this population has the potential to decrease overdose mortality. For every 100 people enrolled in this study, 10 people used naloxone in the six-month follow up period. In over 70% of witnessed events (n = 50), participants administered naloxone successfully despite receiving only a very short training. In other populations, the rate of naloxone use among trained participants ranged from five to 13% at three-months (McCauley et al., 2010), and 16% at six-months after training (Siegler et al., 2017). The naloxone use rate for this study's cohort was nearly as high as among that of active drug users, suggesting that individuals who visit jails are at high risk of witnessing overdose and may be connected to social networks in which naloxone can play an integral role in overdose mortality reduction. In contrast to studies of people trained in ORT at SEPs, the majority of this study population was female (Siegler et al., 2017). This finding highlights an opportunity to reach an untapped social network of individuals who are witnessing drug overdoses in New York City. Our program was feasible to implement and is a novel way to reach people at high risk for observing an overdose. Only three of the witnessed overdoses involved the person visited at Rikers Island on the day of ORT. In an additional nine witnessed events, the overdose victims had been incarcerated (not necessarily at Rikers) within the last year. Nearly half (47%) of overdose victims had unknown incarceration history. Our finding suggests that training visitors of incarcerated individuals may prevent overdose fatalities among a population beyond the recently incarcerated. Our study cannot determine the extent to which this training program reached people who were recently released and subsequently overdosed. Though release from incarceration is an important risk factor for overdose (Binswanger et al., 2007; Lim et al., 2012), the absolute rate of death following release is low with a recent study of the Rikers Island population reporting 22 opioid overdose deaths among over 86,000 recently incarcerated persons (Alex et al., 2017). Although our program reversed overdoses in only three visited individuals, this may represent a substantial portion of visited persons who experienced an overdose within six months of their return to the community. With the mean length of stay at NYC jails of 49 days (MacDonald et al., 2015), some visited persons could have still been incarcerated during the six-months study follow up. Additionally, we do not know what role the visitor has in the incarcerated individual's social network and so they could have overdosed without the visitor's knowledge. The success of this intervention to prevent overdose fatalities among released individuals by training visitors is therefore unknown. Other interventions, such as directly equipping individuals being released from jail, should be explored further. We were not able to assess the impact of the training on knowledge due to our attempt to constrain the time commitment for follow up surveys. Our previous study of similar trainings of syringe exchange participants found high levels of knowledge about overdose prevention and naloxone administration at six months after brief trainings (Maldjian, Siegler, & Kunins, 2016). Additionally, a similar study of brief overdose rescue trainings in San Francisco also demonstrated high levels of knowledge about naloxone use in individuals returning for
4.1. Conclusion In summary, training visitors to incarcerated individuals in overdose rescue is an effective strategy to reach a population of potential overdose responders. Although few participants reversed an overdose of the person they visited on the day of ORT, 10% did respond to an overdose within six months of training. As the opioid overdose epidemic grows, naloxone programs need to continue to equip a wide range of responders. Training jail visitors is one key strategy to reach a social network likely to observe overdose. Acknowledgements This research was supported by Grant No. 2014-PM-BX-0015 awarded by the Bureau of Justice Assistance. The Bureau of Justice Assistance is a component of the Department of Justice's Office of Justice Programs, which also includes the Bureau of Justice Statistics, the National Institute of Justice, the Office of Juvenile Justice and Delinquency Prevention, the Office for Victims of Crime, and the SMART Office. Points of view or opinions in this document are those of the authors and do not necessarily represent the official position or policies of the U.S. Department of Justice. References Alex, B., Weiss, D. B., Kaba, F., Rosner, Z., Lee, D., Lim, S., ... MacDonald, R. (2017). Death after jail release. Journal Correctional Health Care, 23, 83–87. http://dx.doi. org/10.1177/1078345816685311. Drug Policy Alliance. (2015) 911 Good Samaritan: Explaining New York's overdose prevention law. Available at: http://www.drugpolicy.org/sites/default/files/911_Good_ Samaritan_Informational_Brief.pdf. Behar, E., Santos, G. M., Wheeler, E., Rowe, C., & Coffin, P. O. (2015). Brief overdose education is sufficient for naloxone distribution to opioid users. Drug and Alcohol Dependence, 148, 209–212. http://dx.doi.org/10.1016/j.drugalcdep.2014.12.009. Beletsky, L., LaSalle, L., Newman, M., Paré, J., Tam, J. S., & Tochka, A. B. (2015). Fatal reentry: Legal and programmatic opportunities to curb opioid overdose among individuals newly released from incarceration. School of Law Faculty Publications, 7, 155–215. Available at: http://lsr.nellco.org/nusl_faculty/17. Bennett, T., & Halloway, K. (2012). The impact of take-home naloxone distribution and training on opiate overdose knowledge and response: An evaluation of the THN Project in Wales. Drugs: Education, Prevention and Policy, 19, 320–328. http://dx.doi. org/10.3109/09687637.2012.658104. Binswanger, I. A., Stern, M. F., Deyo, R. A., Heagerty, P. J., Cheadle, A., Elmore, J. G., & Koespsell, T. D. (2007). Release from prison - a high risk of death for former inmates. New England Journal of Medicine, 356, 157–165. http://dx.doi.org/10.1056/ NEJMsa064115. Bird, S. M., & Hutchinson, S. J. (2003). Male drugs-related deaths in the fortnight after release from prison: Scotland 1996–99. Addiction, 98, 185–190. http://dx.doi.org/10. 1046/j.1360-0443.2003.00264.x. Bird, S. M., Parmar, M. K., & Strang, J. (2015). Take-home naloxone to prevent fatalities from opiate-overdose: Protocol from Scotland's public health policy evaluation, and a new measure to assess impact. Drugs, 22, 66–76. http://dx.doi.org/10.3109/ 09687637.2014.981509. Coffin, P. O., & Sullivan, S. D. (2013). Cost-effectiveness of distributing naloxone to
5
Addictive Behaviors xxx (xxxx) xxx–xxx
Z. Huxley-Reicher et al.
Journal, 5, 2. http://dx.doi.org/10.1186/1477-7517-5-2. Siegler, A., Huxley-Reicher, Z., Maldjian, L., Jordan, R., Oliver, C., Jakubowski, A., & Kunins, H. V. (2017). Naloxone use among overdose prevention trainees in New York City: A longitudinal cohort study. Drug and Alcohol Dependence, 179, 124–130. http:// dx.doi.org/10.1016/j.drugalcdep.2017.06.029. Stewart, L. M., Henderson, C. J., Hobbs, M. S., Ridout, S. C., & Knuiman, M. W. (2004). Risk of death in prisoners after release from jail. Australian and New Zealand Journal Public Health, 28, 32–36. http://dx.doi.org/10.1111/j.1467-842X.2004.tb00629.x. Strang, J., Manning, V., Mayet, S., Best, D., Titherington, E., Santana, L., et al. (2008). Overdose training and take-home naloxone for opiate users: Prospective cohort study of impact on knowledge and attitude and subsequent management of overdoses. Addiction, 103, 1648–1657. http://dx.doi.org/10.1111/j.1360-0443.2008.02314.x. Tobin, K. E., Sherman, S. G., Beilenson, P., Welsh, C., & Latkin, C. A. (2009). Evaluation of the Staying Alive programme: Training injection drug users to properly administer naloxone and save lives. International Journal of Drug Policy, 20, 131–136. http://dx. doi.org/10.1016/j.drugpo.2008.03.002. Wagner, K. D., Valente, T. W., Casanova, M., Partovi, S. M., Mendenhall, B. M., Hundley, J. H., et al. (2010). Evaluation of an overdose prevention and response training programme for injection drug users in the Skid Row area of Los Angeles, CA. International Journal of Drug Policy, 21, 186–193. http://dx.doi.org/10.1016/j. drugpo.2009.01.003. Walley, A. Y., Xuan, Z., Hackman, H. H., Quinn, E., Doe-Simkins, M., Sorensen-Alawad, A., et al. (2013). Opioid overdose rates and implementation of overdose education and nasal naloxone distribution in Massachusetts: Interrupted time series analysis. BMJ, 346, f174. http://dx.doi.org/10.1136/bmj.f174. Wheeler, E., Jones, T. S., Gilbert, M. K., & Davidson, P. J. (2015). Opioid overdose prevention programs providing naloxone to laypersons – United States, 2014. Morbidity and Mortality Weekly Report, 64, 631–635. Winter, R. J., Stoove, M., Degenhardt, L., Hellard, M. E., Spelman, T., Jenkinson, R., et al. (2015). Incidence and predictors of non-fatal drug overdose after release from prison among people who inject drugs in Queensland, Australia. Drug and Alcohol Dependence, 153, 43–49. http://dx.doi.org/10.1016/j.drugalcdep.2015.06.011.
heroin users for lay overdose reversal. Annals of Internal Medicine, 158, 1–9. http:// dx.doi.org/10.7326/0003-4819-158-1-201301010-00003. Farrell, M., & Marsden, J. (2008). Acute risk of drug-related death among newly released prisoners in England and Wales. Addiction, 103, 251–255. http://dx.doi.org/10. 1111/j.1360-0443.2007.02081.x. Gaston, R. L., Best, D., Manning, V., & Day, E. (2009). Can we prevent drug related deaths by training opioid users to recognise and manage overdoses. Harm Reduction Journal, 6, 26. http://dx.doi.org/10.1186/1477-7517-6-26. Lim, S., Seligson, A. L., Parvez, F. M., Luther, C. W., Mavinkurve, M., Binswanger, I., et al. (2012). Risks of drug-related death, suicide, and homicide during the immediate postrelease period among people released from New York City jails, 2001–2005. American Journal of Epidemiology, 175, 519–526. http://dx.doi.org/10.1093/aje/kwr327. MacDonald, R., Kaba, F., Rosner, Z., Vise, A., Weiss, D., Brittner, W., ... Venters, H. (2015). The Rikers Island hot spotters: Defining the needs of the most frequently incarcerated. American Journal of Public Health, 105, 2262–2268. http://dx.doi.org/ 10.2105/AJPH.2015.302785. Maldjian, L., Siegler, A., & Kunins, H. V. (2016). Evaluation of overdose prevention trainings in New York City: Knowledge and self-efficacy among participants 12 months after training. Substance Abuse, 37(3), 459–465. http://dx.doi.org/10.1080/ 08897077.2015.1135850. McAuley, A., Lindsay, G., Woods, M., & Louttit, D. (2010). Responsible management and use of a personal take-home naloxone supply: A pilot project. Drugs: Education, Prevention and Policy, 17, 388–399. http://dx.doi.org/10.3109/ 09687630802530712. McCauley, A., Aucott, L., & Matheson, C. (2015). Exploring the life-saving potential of naloxone: A systematic review and descriptive meta-analysis of take home naloxone (THN) programmes for opioid users. International Journal of Drug Policy, 26, 1183–1188. http://dx.doi.org/10.1016/j.drugpo.2015.09.011. Merrall, E. L., Kariminia, A., Binswanger, I. A., Hobbs, M. S., Farrell, M., Marsden, J., et al. (2010). Meta-analysis of drug-related deaths soon after release from prison. Addiction, 105, 1545–1554. http://dx.doi.org/10.1111/j.1360-0443.2010.02990.x. Sherman, S. G., Gann, D. S., Scott, G., Carlberg, S., Bigg, D., & Heimer, R. (2008). A qualitative study of overdose responses among Chicago IDUs. Harm Reduction
6
Contents lists available at ScienceDirect
Addictive Behaviors journal homepage: www.elsevier.com/locate/addictbeh
Witnessed overdoses and naloxone use among visitors to Rikers Island jails trained in overdose rescue ⁎
Zina Huxley-Reichera, ,1, Lara Maldjiana, Emily Winkelsteina, Anne Sieglerb, Denise Paonea, Ellenie Tuazona, Michelle L. Nolana, Alison Jordanb, Ross MacDonaldb, Hillary V. Kuninsa a b
New York City Department of Health and Mental Hygiene, 42-09 28th Street, 19th Floor, Long Island City, NY 11101, United States New York City Health + Hospitals, 55 Water Street, 18th floor, New York, NY 10014, United States
H I G H L I G H T S prospective study of NYC jail visitors to Rikers Island trained in naloxone. • AOf6-month 283 participants enrolled, 14% witnessed at least one overdose. • Of the 283 participants enrolled, 10% administered naloxone at least once. • Thethenaloxone use is comparable to similar interventions for high-risk populations. • Training jail visitors is effective at reaching a population at risk of overdose. •
A R T I C L E I N F O
A B S T R A C T
Keywords: Naloxone Jail Overdose Rescue Criminal justice Opioid
With the opioid overdose mortality rates rising nationally, The New York City Department of Health and Mental Hygiene (NYC DOHMH) has worked to expand overdose rescue training (ORT) and naloxone distribution. This study sought to determine rates of overdose witnessing and naloxone use among overdose rescue-trained visitors to the NYC jails on Rikers Island. We conducted a six-month prospective study of visitors to NYC jails on Rikers Island who received ORT. We collected baseline characteristics of study participants, characteristics of overdose events, and responses to witnessed overdose events, including whether the victim was the incarcerated individual the participant was visiting on the day of training. Bivariate analyses compared baseline characteristics of participants who witnessed overdoses to those who did not, and of participants who used naloxone to those who did not. Overall, we enrolled 283 participants visiting NYC's Rikers Island jails into the study. Six months after enrollment, we reached 226 participants for follow-up by phone. 40 participants witnessed 70 overdose events, and 28 participants reported using naloxone. Of the 70 overdose events, three victims were the incarcerated individuals visited on the day of training; nine additional victims were recently released from jail and/or prison. Visitors to persons incarcerated at Rikers Island witness overdose events and are able to perform overdose rescues with naloxone. This intervention reaches a population that includes not only those recently released, but also other people who experienced overdose.
1. Introduction Mortality rates from unintentional opioid overdose continue to increase across the United States. Between 2000 and 2014, the rate of opioid overdose deaths tripled (Wheeler, Jones, Gilbert, & Davidson, 2015). Opioid overdose deaths are preventable with a multi-pronged approach, including naloxone distribution to laypeople. Naloxone is an opioid antagonist medication that reverses the effect of opioid intoxication (overdose) and restores breathing. In New York State, opioid ⁎
1
overdose prevention programs, like overdose prevention programs in other jurisdictions, educate laypeople to recognize and respond to an overdose, including how to administer naloxone. Distribution of naloxone to laypeople at risk of overdose is cost-effective and reduces overdose mortality (Coffin & Sullivan, 2013; Gaston, Best, Manning, & Day, 2009; Sherman et al., 2008; Strang et al., 2008; Wagner et al., 2010; Walley et al., 2013). In the United States, there have been over 26,000 reversals reported and over 150,000 kits dispensed since 1996 (Wheeler et al., 2015).
Corresponding author. E-mail address: [email protected] (Z. Huxley-Reicher). Present address: Icahn School of Medicine at Mount Sinai; 1 Gustave L Levy Place, New York, NY 10029, United States.
https://doi.org/10.1016/j.addbeh.2017.11.029 Received 25 July 2017; Received in revised form 6 November 2017; Accepted 15 November 2017 0306-4603/ © 2017 Published by Elsevier Ltd.
Please cite this article as: Huxley-Reicher, Z., Addictive Behaviors (2017), https://doi.org/10.1016/j.addbeh.2017.11.029
Addictive Behaviors xxx (xxxx) xxx–xxx
Z. Huxley-Reicher et al.
verbal consent which included education about the study's risks and benefits as well as confidentiality and study procedures. All trained individuals received naloxone kits as they exited Rikers Island Central Visit Center, and after completing their visit to the incarcerated person. Study participants included all individuals who 1) participated in ORT, 2) gave verbal consent to participate in the study, 3) provided demographic (age, race/ethnicity and gender) and contact information in either Spanish or English, and 4) received naloxone after their visit. IRBs at the NYC DOHMH and NYC Health and Hospitals approved the study.
In order for naloxone distribution to reduce overdose mortality effectively, distribution to populations at highest risk of overdose should be prioritized. In many jurisdictions, including New York City (NYC), syringe exchange programs (SEPs) accomplish this goal by engaging and training people who actively use drugs to recognize and respond to an overdose, including how to administer naloxone. A meta-analysis of community naloxone trainings among people who use drugs reported naloxone use rates that ranged from 5 to 13% in the three months after naloxone receipt (McCauley, Aucott, & Matheson, 2015). NYC SEP participants trained to use naloxone reported higher rates; 25% used naloxone to reverse an overdose within 12 months of training (Siegler et al., 2017). Following release from carceral settings, individuals are at risk of fatal overdose due to reduced opioid tolerance after a period of abstinence. In NYC, people with a recent history of incarceration are eight times more likely to die from drug-related causes in the two weeks after incarceration than non-incarcerated individuals (Lim et al., 2012). Similar findings are reported in other parts of the United States and internationally (Binswanger et al., 2007; Farrell & Marsden, 2008; Bird and Hutchinson, 2003; Stewart, Henderson, Hobbs, Ridout, & Knuiman, 2004; Merrall et al., 2010; Winter et al., 2015). Additionally, of the > 200,000 individuals released from NYC jails between 2011 and 2016, 36,000 (18%) had an opioid use disorder (unpublished data, Correctional Health Service, NYC Health and Hospitals). In response to the high risk of overdose following incarceration, a number of programs dispense naloxone to formerly incarcerated persons upon re-entry. One program in Scotland found reductions in overdose mortality rates following the implementation of naloxone dispensing to people at prison discharge (Bird, Parmar, & Strang, 2015). In the U.S., programs in Rhode Island, California, Washington, New Mexico and New York State have begun to provide naloxone to releases (Beletsky et al., 2015). In the NYC jail system, like other correctional systems, a number of logistical barriers have thus far precluded naloxone distribution to incarcerated people upon release. By training visitors, we sought to prevent fatal overdose among persons recently released from jail on Rikers Island, where approximately three-quarters of NYC's jails are located. Beginning in 2014, the New York City Health Department began a program to offer training to visitors in overdose recognition and naloxone administration prior to their jail visit. Those trained received a naloxone kit after their visit, as they exited the Rikers Island Central Visit Center. To determine whether naloxone distribution to jail visitors would lead to overdose rescues among recently incarcerated people, we conducted a prospective study of: 1) the prevalence of witnessing overdose among jail visitors in the 6 months following the visit; 2) naloxone use rates by jail visitors; and 3) the proportion of naloxone administrations by jail visitors that were to individuals recently incarcerated at Rikers Island.
2.2. Overdose rescue training and naloxone dispensing NYC Health Department employees or syringe-exchange outreach workers conducted ORT. ORT lasted from two to five minutes and included instruction on signs of opioid overdose, steps for responding to opioid overdose how to administer intranasal naloxone, and protections provided by the 911 Good Samaritan Law—a New York State law that protects individuals from prosecution when calling 911 in case of an overdose (Drug Policy Alliance, 2015). If time permitted, trainers also instructed participants in rescue breathing and overdose prevention tips. Due to restrictions on property permitted inside individual jail facilities, staff gave naloxone kits to trained visitors after their visit, as they exited the jail complex, often three to four hours after initial training. Each naloxone kit contained two doses of 1 mg/mL naloxone with two nasal adaptors, disposable gloves, an alcohol swab, instructions for use of the product, and a rescue breathing mask. 2.3. Follow up procedures We asked study participants to provide at least one mechanism for follow up contact (phone number or email address) and a mailing address to receive incentives. Participants were reached by phone call or email to confirm contact information at one month and three months after enrollment. At six months, we administered a closed-ended survey by phone. Participants who agreed to be contacted via text message also received texts each month to maintain contact. At one, three, and six months, we attempted to reach participants by phone-call up to three times, and by email and post card if unsuccessful by phone. 2.4. Incentives Following enrollment at the Central Visit Center, we gave participants round-trip subway cards valued at $5.50. Participants received an additional subway card upon completion of the one-month check-in, two subway cards upon completion of the three month check in, and a $20 gift card upon completion of the six-month follow-up survey. The maximum incentive was $42. 2.5. Measures
2. Methods We collected demographic information immediately following ORT. Demographic characteristics of study participants included age, gender, race/ethnicity, and borough of residence. The questions on the sixmonth survey included questions about witnessing overdose, use of naloxone, incarceration history of overdose victim, and relationship to overdose victim.
2.1. Sample and study enrollment We conducted a prospective observational study of individuals visiting incarcerated persons at Rikers Island who completed overdose rescue training (ORT) and received naloxone. Participants were recruited by convenience sampling from the Rikers Island Central Visit Center during five consecutive days in August 2015. These represented all days available for visiting in a week (not every day is open and days are divided by last name of the individual being visited). Upon entry into the Rikers Island Central Visit Center, program staff offered all visitors ORT, regardless of the sentencing status of the incarcerated person they were visiting, while they awaited transport to the individual jails in the complex. Staff invited individuals who agreed to be trained to participate in the study. Staff then obtained informed
2.6. Outcomes The three outcomes of interest were: (1) witnessing a drug overdose, (2) administering naloxone, and (3) whether the overdose victims were the individuals being visited at Rikers on the day of training. 2.6.1. Witnessing drug overdose At six months, participants reported whether they had witnessed 2
Addictive Behaviors xxx (xxxx) xxx–xxx
Z. Huxley-Reicher et al.
Fig. 1. Flow chart of participants in Rikers Island visitor overdose rescue training.
ORT and naloxone administration during any witnessed overdose event dichotomously. To calculate the rate of witnessed overdoses, the number of participants who witnessed an overdose in the six-month evaluation period was divided by the total number of study participants. We calculated the rate of naloxone administrations by dividing the number of participants who used naloxone in a six-month time period by the total number of study participants.
any overdoses since study enrollment. An overdose was defined to study participants as “unresponsive or unable to be woken up, collapsing, having blue skin color, having difficulty breathing, losing consciousness or dying while using drugs.” For each witnessed overdose, participants reported (1) their relationship to victim, (2) survival status of victim, (3) recent incarceration status of victim (whether they had been in jail or prison in the last year), (4) and their responses to the event. 2.6.2. Naloxone administration Participants reported naloxone administration for each witnessed overdose. For overdoses where a layperson administered naloxone, participants indicated whether they or another bystander administered naloxone. If an overdose was witnessed, but naloxone was not administered, we asked the main reason it was not administered. Participants who did administer naloxone reported whether they observed a response in the victim after the naloxone administration. A secondary outcome assessed whether participants or another bystander called 911. We also asked participants if they obtained a replacement naloxone kit following administration.
2.7.2. Event-level variables We categorized whether the overdose victim was the person visited at the time of ORT training dichotomously. We categorized past-year incarceration of the overdose victim trichotomously as yes, no or don't know. Relationship to overdose victim included four categories: friend, family member, acquaintance and stranger. We treated survival status as assessed by participants and whether 911-calling as dichotomous variables. Descriptive analyses of study participants and overdose events were conducted. Using univariate logistic regression, we calculated 95% confidence intervals and p-values to determine which variables (age, gender, race/ethnicity, day of enrollment, and borough of residence) were associated with witnessing overdose and administering naloxone. The small number of participants who reported witnessing an overdose where the victim was the individual visited on the day of training precluded bivariable analysis. Statistical significance was defined as p < 0.05; analyses were performed using SAS 9.2 (Cary, North Carolina).
2.6.3. Overdose victim's incarceration history For each witnessed overdose event, participants reported whether the victim was the individual they were visiting on the day they were trained at Rikers. If not, research staff asked whether the victim had been recently released (within the last year) from jail or prison. 2.7. Data analysis We compared demographic differences between those who completed and did not complete the six-month survey using a chi-square to determine whether loss to follow-up could have biased study findings. The analyses included both individual (by participant) and the event (by overdose event) levels.
3. Results During the five days of enrollment, 382 visitors were trained in overdose rescue at the Rikers Island Visit Center. A total of 335 (88%) returned to request a naloxone kit after their visit, and 283 (85%) consented to participate in the study. A total of 226 participants who consented to participate (80%) were reached and completed the sixmonth follow up (Fig. 1). Our study population was 83.2% female, 43.6% African American, 38.2% Hispanic and the mean age was 38.4 years old. We reached younger visitors, aged 16–24 years, less frequently to complete the 6-month follow-up survey compared to the sample as a whole (69% vs 80%, p < 0.04).
2.7.1. Individual-level variables We collapsed age (years) into five categories, 16–24, 25–34, 35–44, 45–54 and 55 and older. We categorized race/ethnicity as Hispanic/ Latino, non-Hispanic Black/African American, non-Hispanic White, and other. We categorized enrollment date as weekday or weekend. We categorized witnessing at least one overdose in the six months following 3
Addictive Behaviors xxx (xxxx) xxx–xxx
Z. Huxley-Reicher et al.
Table 1 Characteristics of participants who completed overdose rescue training (ORT) and completed the six-month survey.
Study population (n = 283) Age (mean) 16–24 25–34 35–44 45–54 55 + Gendera Male Female Race/ethnicity Hispanic/Latino Non-Hispanic White Non-Hispanic Black/African American Otherc Borough of residenced Bronx Brooklyn Manhattan Queens Staten Island Other Day of overdose rescue training Weekday Weekend Mean number of witnessed overdoses (range)
Total sample who completed 6 month survey
Witnessed an overdosea
Naloxone administeredb
n (%e)
n (%f)
n (%g)
226 (79.9%)
40 (14.1)
28 (9.9)
38.4 40 (17.7) 58 (25.7) 52 (23.0) 40 (17.7) 36 (15.9)
40.9 5 (12.5) 8 (20.0) 11 (27.5) 6 (15.0) 10 (25.0)
40.8 4 (14.3) 5 (17.9) 9 (32.1) 2 (7.1) 8 (28.6)
38 (16.8) 188 (83.2)
7 (17.5) 33 (82.5)
5 (17.9) 23 (82.1)
86 (38.2) 13 (5.8)
17 (43.6) 3 (7.7)
13 (48.2) 1 (3.7)
98 (43.6)
18 (46.2)
13 (48.2)
28 (12.4)
1 (2.6)
0
60 55 31 37 21 22
12 (30.0) 10 (25.0) 6 (15.0) 4 (10.0) 4 (10.0) 4 (10.0)
11 (39.3) 7 (25.0) 5 (17.9) 0 2 (7.1) 3 (10.7)
30 (75.0) 10 (25.0) 1.8(1–4)
20 (71.4) 8 (29.6) 1.9 (1-4h)
(26.6) (24.3) (13.7) (16.4) (9.3) (9.7)
170 (75.2) 56 (24.8) 0.3 (0–4)
Table 2 Characteristics of overdose events. Event characteristics
Number of events (%)
Total events Victim characteristics Recently released from jail or prisona Yes No Don't Know Was visited by study participant on day of naloxone receipt Relationship Familyb Friend Stranger Acquaintancec Survival Yes No Do not know Responses 911 called by anyone Naloxone given by participant by another layperson
70 (100)
12 (17.1) 25 (35.7) 33 (47.1) 3 (4.3)
12 14 32 12
(17.1) (20) (45.7) (17.1)
61 (87.1) 4 (5.7%) 5 (7.1%) 56 (80) 50 (71.4) 5 (7.1)
a Recently is defined as within the last year (n = 6 events were < 6 months after release, n = 4 events were > 6 months after release and n = 2 were unknown). b Family includes: victim's parent, sibling, relative, and spouse/partner. c Acquaintance includes: client, neighbor and acquaintance.
relevant percentages but not the odds ratios or p values).
3.2. Naloxone administration During the 70 witnessed overdose events, a lay person administered naloxone in 55 events (79%); study participants, specifically, administered naloxone in 71% of events. Overall, 28 (10%) study participants reported administering naloxone at least once during the study period. Either a participant or another bystander called 911 in 56 (80%) of the witnessed overdose events (Table 2). The main reasons that participants reported for not administering naloxone at overdose events (n = 20, 29%) were: emergency personnel arrived first (n = 8; 40%), the participant did not have naloxone on hand (n = 5; 25%), and someone else administered naloxone (n = 3; 15%) (Table 3). None of the measured participant characteristics were associated with administering naloxone (see Table 1). Of the 55 overdoses where lay people administered naloxone, participants reported that the victim responded to the naloxone (“woke up after they were given naloxone”) in 50 events (91%). Of the four events where the victim did not respond to naloxone administration, the victim did not survive in three, and had unknown survival in one. In one event, the participant did not answer the question about victim's response to naloxone administration.
a
No variables were significantly associated with witnessing an overdose. No variables were significantly associated with using naloxone. c Includes those who reported Asian, Native American, Mixed Race. d Residence at the time of recruitment. e Column %- denominator is the total n = 226 for those who completed the 6 month survey. f Column % - denominator is the total n = 40 for those who have witness an OD. g Column % - denominator is the total n = 28 of those who administered Naloxone. h All individuals who witnessed 3 (n = 6) and 4 (n = 3) ODs used naloxone in at least one of the witnessed OD events. b
3.1. Witnessing drug overdose Of the 226 participants reached at six months, 40 (14% of the total enrolled n = 283) had witnessed at least one overdose during the study period (Table 1). Twenty-three (23, 56%) participants witnessed one overdose, nine (23%) witnessed two, five (13%) witnessed three and three (8%) participants witnessed four events during the six months following ORT. In total, study participants witnessed 70 overdose events. Survival data was available for 65 of the 70 events (92%) (Table 2); of those 65 events, the victim survived in 61. The relationship between the participant and victim varied: 46% of the victims were strangers to the participant, 20% were friends, 17% were family members, and 17% were acquaintances. In 47% of events the victim's incarceration history was unknown; in 17% of events, victims were recently released from jail or prison. In three (4%) of the events, the overdose victim was the person visited on the day the participant received ORT. No variables measured (age, gender, race/ethnicity, borough of residence, or weekday/weekend enrollment) were associated with witnessing an overdose (see Table 1; to conserve space, we show the
Table 3 Main reason naloxone was not administered by participant at witnessed OD events. Reason
n (%)
Total events Emergency personnel arrived first Didn't have naloxone on hand Someone else administered naloxonea Other Had already used kit
20 (100) 8 (40) 5 (25) 3 (15) 3 (15) 1 (5)
a In an additional 2 events naloxone was administered by another bystander; but this was not the main reason the participant didn't administer.
4
Addictive Behaviors xxx (xxxx) xxx–xxx
Z. Huxley-Reicher et al.
4. Discussion
naloxone refills after receipt of brief overdose rescue training (Behar, Santos, Wheeler, Rowe, & Coffin, 2015). This study had additional limitations. Due to the restrictions on what can be brought into jail facilities, we distributed naloxone hours after the initial training and 12% of those trained did not receive a kit. In order to minimize participants' burden, we limited the number of questions on the six-month survey. This resulted in limited data about the witnessed events and the use of naloxone. Also, we found no associations of independent variables with our outcomes (witnessing overdose and using naloxone). While this may demonstrate a real pattern, it is likely due to the fact that our study was not powered sufficiently to detect differences. In addition, 20% of the initial study population were lost to follow up; it is possible that participants who responded at six-months were more likely to witness and respond to overdose, thereby biasing our results upwards. By including the rate among the total study population, rather than the responders only, we sought to minimize this bias.
Visitors to the Rikers Island Central Visit Center who received ORT had a high frequency of witnessing overdoses post-training. Although this intervention successfully reached a population that witnessed overdoses in rates similar to other overdose rescue training programs, few of the overdoses witnessed were among people visited on the day of training. Study participants witnessed overdoses at nearly the same rate reported in previous studies of naloxone training programs (Bennett & Halloway, 2012; McAuley, Lindsay, Woods, & Louttit, 2010; Strang et al., 2008; Tobin, Sherman, Beilenson, Welsh, & Latkin, 2009). Our finding that visitors of individuals incarcerated at Rikers Island who are trained in ORT witness and respond to overdoses at high rates demonstrates that distributing naloxone to this population has the potential to decrease overdose mortality. For every 100 people enrolled in this study, 10 people used naloxone in the six-month follow up period. In over 70% of witnessed events (n = 50), participants administered naloxone successfully despite receiving only a very short training. In other populations, the rate of naloxone use among trained participants ranged from five to 13% at three-months (McCauley et al., 2010), and 16% at six-months after training (Siegler et al., 2017). The naloxone use rate for this study's cohort was nearly as high as among that of active drug users, suggesting that individuals who visit jails are at high risk of witnessing overdose and may be connected to social networks in which naloxone can play an integral role in overdose mortality reduction. In contrast to studies of people trained in ORT at SEPs, the majority of this study population was female (Siegler et al., 2017). This finding highlights an opportunity to reach an untapped social network of individuals who are witnessing drug overdoses in New York City. Our program was feasible to implement and is a novel way to reach people at high risk for observing an overdose. Only three of the witnessed overdoses involved the person visited at Rikers Island on the day of ORT. In an additional nine witnessed events, the overdose victims had been incarcerated (not necessarily at Rikers) within the last year. Nearly half (47%) of overdose victims had unknown incarceration history. Our finding suggests that training visitors of incarcerated individuals may prevent overdose fatalities among a population beyond the recently incarcerated. Our study cannot determine the extent to which this training program reached people who were recently released and subsequently overdosed. Though release from incarceration is an important risk factor for overdose (Binswanger et al., 2007; Lim et al., 2012), the absolute rate of death following release is low with a recent study of the Rikers Island population reporting 22 opioid overdose deaths among over 86,000 recently incarcerated persons (Alex et al., 2017). Although our program reversed overdoses in only three visited individuals, this may represent a substantial portion of visited persons who experienced an overdose within six months of their return to the community. With the mean length of stay at NYC jails of 49 days (MacDonald et al., 2015), some visited persons could have still been incarcerated during the six-months study follow up. Additionally, we do not know what role the visitor has in the incarcerated individual's social network and so they could have overdosed without the visitor's knowledge. The success of this intervention to prevent overdose fatalities among released individuals by training visitors is therefore unknown. Other interventions, such as directly equipping individuals being released from jail, should be explored further. We were not able to assess the impact of the training on knowledge due to our attempt to constrain the time commitment for follow up surveys. Our previous study of similar trainings of syringe exchange participants found high levels of knowledge about overdose prevention and naloxone administration at six months after brief trainings (Maldjian, Siegler, & Kunins, 2016). Additionally, a similar study of brief overdose rescue trainings in San Francisco also demonstrated high levels of knowledge about naloxone use in individuals returning for
4.1. Conclusion In summary, training visitors to incarcerated individuals in overdose rescue is an effective strategy to reach a population of potential overdose responders. Although few participants reversed an overdose of the person they visited on the day of ORT, 10% did respond to an overdose within six months of training. As the opioid overdose epidemic grows, naloxone programs need to continue to equip a wide range of responders. Training jail visitors is one key strategy to reach a social network likely to observe overdose. Acknowledgements This research was supported by Grant No. 2014-PM-BX-0015 awarded by the Bureau of Justice Assistance. The Bureau of Justice Assistance is a component of the Department of Justice's Office of Justice Programs, which also includes the Bureau of Justice Statistics, the National Institute of Justice, the Office of Juvenile Justice and Delinquency Prevention, the Office for Victims of Crime, and the SMART Office. Points of view or opinions in this document are those of the authors and do not necessarily represent the official position or policies of the U.S. Department of Justice. References Alex, B., Weiss, D. B., Kaba, F., Rosner, Z., Lee, D., Lim, S., ... MacDonald, R. (2017). Death after jail release. Journal Correctional Health Care, 23, 83–87. http://dx.doi. org/10.1177/1078345816685311. Drug Policy Alliance. (2015) 911 Good Samaritan: Explaining New York's overdose prevention law. Available at: http://www.drugpolicy.org/sites/default/files/911_Good_ Samaritan_Informational_Brief.pdf. Behar, E., Santos, G. M., Wheeler, E., Rowe, C., & Coffin, P. O. (2015). Brief overdose education is sufficient for naloxone distribution to opioid users. Drug and Alcohol Dependence, 148, 209–212. http://dx.doi.org/10.1016/j.drugalcdep.2014.12.009. Beletsky, L., LaSalle, L., Newman, M., Paré, J., Tam, J. S., & Tochka, A. B. (2015). Fatal reentry: Legal and programmatic opportunities to curb opioid overdose among individuals newly released from incarceration. School of Law Faculty Publications, 7, 155–215. Available at: http://lsr.nellco.org/nusl_faculty/17. Bennett, T., & Halloway, K. (2012). The impact of take-home naloxone distribution and training on opiate overdose knowledge and response: An evaluation of the THN Project in Wales. Drugs: Education, Prevention and Policy, 19, 320–328. http://dx.doi. org/10.3109/09687637.2012.658104. Binswanger, I. A., Stern, M. F., Deyo, R. A., Heagerty, P. J., Cheadle, A., Elmore, J. G., & Koespsell, T. D. (2007). Release from prison - a high risk of death for former inmates. New England Journal of Medicine, 356, 157–165. http://dx.doi.org/10.1056/ NEJMsa064115. Bird, S. M., & Hutchinson, S. J. (2003). Male drugs-related deaths in the fortnight after release from prison: Scotland 1996–99. Addiction, 98, 185–190. http://dx.doi.org/10. 1046/j.1360-0443.2003.00264.x. Bird, S. M., Parmar, M. K., & Strang, J. (2015). Take-home naloxone to prevent fatalities from opiate-overdose: Protocol from Scotland's public health policy evaluation, and a new measure to assess impact. Drugs, 22, 66–76. http://dx.doi.org/10.3109/ 09687637.2014.981509. Coffin, P. O., & Sullivan, S. D. (2013). Cost-effectiveness of distributing naloxone to
5
Addictive Behaviors xxx (xxxx) xxx–xxx
Z. Huxley-Reicher et al.
Journal, 5, 2. http://dx.doi.org/10.1186/1477-7517-5-2. Siegler, A., Huxley-Reicher, Z., Maldjian, L., Jordan, R., Oliver, C., Jakubowski, A., & Kunins, H. V. (2017). Naloxone use among overdose prevention trainees in New York City: A longitudinal cohort study. Drug and Alcohol Dependence, 179, 124–130. http:// dx.doi.org/10.1016/j.drugalcdep.2017.06.029. Stewart, L. M., Henderson, C. J., Hobbs, M. S., Ridout, S. C., & Knuiman, M. W. (2004). Risk of death in prisoners after release from jail. Australian and New Zealand Journal Public Health, 28, 32–36. http://dx.doi.org/10.1111/j.1467-842X.2004.tb00629.x. Strang, J., Manning, V., Mayet, S., Best, D., Titherington, E., Santana, L., et al. (2008). Overdose training and take-home naloxone for opiate users: Prospective cohort study of impact on knowledge and attitude and subsequent management of overdoses. Addiction, 103, 1648–1657. http://dx.doi.org/10.1111/j.1360-0443.2008.02314.x. Tobin, K. E., Sherman, S. G., Beilenson, P., Welsh, C., & Latkin, C. A. (2009). Evaluation of the Staying Alive programme: Training injection drug users to properly administer naloxone and save lives. International Journal of Drug Policy, 20, 131–136. http://dx. doi.org/10.1016/j.drugpo.2008.03.002. Wagner, K. D., Valente, T. W., Casanova, M., Partovi, S. M., Mendenhall, B. M., Hundley, J. H., et al. (2010). Evaluation of an overdose prevention and response training programme for injection drug users in the Skid Row area of Los Angeles, CA. International Journal of Drug Policy, 21, 186–193. http://dx.doi.org/10.1016/j. drugpo.2009.01.003. Walley, A. Y., Xuan, Z., Hackman, H. H., Quinn, E., Doe-Simkins, M., Sorensen-Alawad, A., et al. (2013). Opioid overdose rates and implementation of overdose education and nasal naloxone distribution in Massachusetts: Interrupted time series analysis. BMJ, 346, f174. http://dx.doi.org/10.1136/bmj.f174. Wheeler, E., Jones, T. S., Gilbert, M. K., & Davidson, P. J. (2015). Opioid overdose prevention programs providing naloxone to laypersons – United States, 2014. Morbidity and Mortality Weekly Report, 64, 631–635. Winter, R. J., Stoove, M., Degenhardt, L., Hellard, M. E., Spelman, T., Jenkinson, R., et al. (2015). Incidence and predictors of non-fatal drug overdose after release from prison among people who inject drugs in Queensland, Australia. Drug and Alcohol Dependence, 153, 43–49. http://dx.doi.org/10.1016/j.drugalcdep.2015.06.011.
heroin users for lay overdose reversal. Annals of Internal Medicine, 158, 1–9. http:// dx.doi.org/10.7326/0003-4819-158-1-201301010-00003. Farrell, M., & Marsden, J. (2008). Acute risk of drug-related death among newly released prisoners in England and Wales. Addiction, 103, 251–255. http://dx.doi.org/10. 1111/j.1360-0443.2007.02081.x. Gaston, R. L., Best, D., Manning, V., & Day, E. (2009). Can we prevent drug related deaths by training opioid users to recognise and manage overdoses. Harm Reduction Journal, 6, 26. http://dx.doi.org/10.1186/1477-7517-6-26. Lim, S., Seligson, A. L., Parvez, F. M., Luther, C. W., Mavinkurve, M., Binswanger, I., et al. (2012). Risks of drug-related death, suicide, and homicide during the immediate postrelease period among people released from New York City jails, 2001–2005. American Journal of Epidemiology, 175, 519–526. http://dx.doi.org/10.1093/aje/kwr327. MacDonald, R., Kaba, F., Rosner, Z., Vise, A., Weiss, D., Brittner, W., ... Venters, H. (2015). The Rikers Island hot spotters: Defining the needs of the most frequently incarcerated. American Journal of Public Health, 105, 2262–2268. http://dx.doi.org/ 10.2105/AJPH.2015.302785. Maldjian, L., Siegler, A., & Kunins, H. V. (2016). Evaluation of overdose prevention trainings in New York City: Knowledge and self-efficacy among participants 12 months after training. Substance Abuse, 37(3), 459–465. http://dx.doi.org/10.1080/ 08897077.2015.1135850. McAuley, A., Lindsay, G., Woods, M., & Louttit, D. (2010). Responsible management and use of a personal take-home naloxone supply: A pilot project. Drugs: Education, Prevention and Policy, 17, 388–399. http://dx.doi.org/10.3109/ 09687630802530712. McCauley, A., Aucott, L., & Matheson, C. (2015). Exploring the life-saving potential of naloxone: A systematic review and descriptive meta-analysis of take home naloxone (THN) programmes for opioid users. International Journal of Drug Policy, 26, 1183–1188. http://dx.doi.org/10.1016/j.drugpo.2015.09.011. Merrall, E. L., Kariminia, A., Binswanger, I. A., Hobbs, M. S., Farrell, M., Marsden, J., et al. (2010). Meta-analysis of drug-related deaths soon after release from prison. Addiction, 105, 1545–1554. http://dx.doi.org/10.1111/j.1360-0443.2010.02990.x. Sherman, S. G., Gann, D. S., Scott, G., Carlberg, S., Bigg, D., & Heimer, R. (2008). A qualitative study of overdose responses among Chicago IDUs. Harm Reduction
6