Article, Emergency Medicine

Urban overdose hotspots: a 12-month prospective study in Dublin ambulance services

a b s t r a c t

Background: opioid overdose (OD) is the primary cause of death among drug users globally. Personal and social determinants of overdose have been studied before, but the Environmental factors lacked research attention. Area deprivation or presence of addiction clinics may contribute to overdose.

Objectives: The objective of the study is to examine the baseline incidence of all new ODs in an ambulance service and their relationship with urban deprivation and presence of addiction services.

Methods: A prospective chart review of Prehospital advanced life support patients was performed on confirmed OD calls. Demographic, geographic, and clinical information, that is, presentation, treatment, and outcomes, was collected for each call. The census data were used to calculate deprivation. Geographical information software mapped the urban deprivation and addiction services against the overdose locations. Results: There were 469 overdoses, 13 of which were fatal; most were male (80%), of a young age (32 years), with a high rate of repeated overdoses (26%) and common polydrug use (9.6%). Most occurred in daytime

(275) and on the streets (212). Overdoses were more likely in more affluent areas (r = .15; P b .05) and in a 1000-m radius of addiction services. Residential overdoses were in more deprived areas than street overdoses (mean difference, 7.8; t170 = 3.99; P b .001). Street overdoses were more common in the city center than suburbs (?2(1) = 33.04; P b .001).

Conclusions: The identified clusters of increased incidence–urban overdose hotspots–suggest a link between environment characteristics and overdoses. This highlights a need to establish overdose education and naloxone distribution in the overdose hotspots.

(C) 2014

Introduction

Opioid overdose (OD) is the primary cause of death among drug users globally [1]. Despite an international trend of decreasing drug- related deaths in recent years, opioids remain the major cause of deaths in Ireland [2]. In European countries with widespread Heroin use, such as Ireland, opioids are implicated in 75% of the drug-related deaths [3-5]. In 2011, at least 251 deaths occurred from poisoning with opioids among drug misusers in Ireland; this represents 69% of poisoning deaths in this population [6]. Drug overdose shortens the Life expectancy of drug users compared with the general population. Most drug users witness an overdose, and many are victims of overdose themselves [7-9]. Risk of fatal drug overdose is higher immediately after release from prison or after opioid substitution treatment [10,11]. Acute OD poses a significant burden on frontline services, ambulance services, and emergency departments (ED), but limited evidence is available on the experience or role of those

* Corresponding author. Coombe Family Practice, Dolphin’s Barn, Dublin 8, Ireland. Tel.: +353 14730893; fax: +353 14544469.

E-mail address: [email protected] (J. Klimas).

services [12,13]. Understanding the risk factors and determinants of overdose is critical to help decrease the mortality, morbidity, and burden on health care and to maximize the potential contribution of emergency services.

Drugs are used in a context of social, personal, and environmental characteristics [14]. Contextual effects, such as drug, set, and setting, synergize and create higher chances for overdose [15]. The set represents individual risk factors, which have been well studied: male sex, age, long-term drug use, psychiatric illness, or transition to/from opioid substitution treatment [11]. The setting of drug use includes such understudied environmental factors as income distribution, family fragmentation, physical characteristics of urban areas (clean sidewalks or dilapidated houses), education, or allocation of health services [16-18]. All of these mediate the relationship between the setting and drug use. Building on this evidence base, this study examines 2 additional features of urban areas that could be related to ODs: deprivation and presence of addiction services, in the context of gathering incidence data on overdoses.

First, overdoses and overdose deaths are more frequent in areas of increased drug use and poverty [11]. For example, areas with unequal income distribution have higher rates of overdose deaths, indepen- dent of individual risk factors, such as sex or age [18].

http://dx.doi.org/10.1016/j.ajem.2014.07.017

0735-6757/(C) 2014

Second, areas around addiction services are historically long- established epicenters of drug use.

Addiction services have been traditionally set up in areas of high need [19], where drug dealing, using, and overdosing were already present or dealers may have been attracted to these areas, envisioning higher profits from drug sales and, subsequently, increasing the rates of use and overdose. The evidence from alcohol and tobacco research reveals a link between allocation of outlets and prevalence of tobacco or alcohol use [20]. This relationship has not been studied for addiction clinics or methadone services. Nevertheless, the experience of community-based methadone services suggests that up to 40% of people in recognized methadone treatment continue to inject heroin and may, therefore, bring drug selling and illicit use into the areas around drug treatment centers [7]. This increases the risk of fatal overdose. Moreover, methadone dispensing may have an effect on overdoses, as shown by a 32% increase in overdoses observed in Accidents and Emergency (A&E) in the 2-month period immediately after the introduction of a new methadone prescribing legislation in Ireland, with a subsequent drop (47%) 2 months later [13].

Very little data have been published on the experience of OD by emergency ambulance services, in Ireland or elsewhere. Merchant et al

[21] published a 1997-2002 time series of such events from Rhode Island, in which they reported 1630 events evenly distributed over the period with most being males younger than 54 years and most events occurring in private homes. A 2002 Australian time series indicated an initial drop in the numbers of fatal ODs after a community overdose initiative, but the fall was not sustained [22]. In Dublin, emergency ambulance services are provided by Dublin Fire Brigade (DFB) and the Health Service Executive’s National Ambulance Service (NAS), covering different geographic areas. Emergency ambulances are staffed by paramedics or advanced paramedics who have, since 2004, been trained and equipped to provide naloxone (standard dose 0.4 mg, repeated up to 5 times) in suspected OD using a standardized Clinical Practice Guideline [23].

This study had 2 aims. First, we established baseline incidence of all new overdoses reported to the ambulance services in Dublin over a 12-month period together with their presentation, treatment, and outcomes if known to the ambulance service. Second, we examined the relationship between geographical distribution of overdoses, deprivation, and presence of addiction clinics.

Methods

Setting

Two agencies provide ambulance services in Dublin: DFB and the Health Service Executive, NAS. Established in 1862, DFB is among the oldest ambulance services in the world. It attends to approximately 72000 incidents in Dublin annually. In 2011, naloxone figured in 281 (4.2%) of all medication administrations [24]. National Ambulance Service attends to emergency calls in certain areas of Dublin and dealt with approximately 300000 emergency calls nationally in 2013 [25].

Data collection

Characteristics and clinical data for all ODs reported to DFB and NAS, Dublin, in a 12-month period were collected by the ambulance staff on Patient care report (PCR) forms. Patient care reports are completed in all cases by ambulance service staff, and copies are kept both in the receiving hospital and by the service itself. The study identified all PCRs indicating OD; trained DFB staff collected study data from DFB forms, and the researchers collected data from PCRs in NAS. Population data were derived from the 2011 census.

Patient care report review

Patient care report is a paper-based system for recording prehospital care, assessment observations, interventions, and medi-

cations administered to patients by the emergency responders. Hard copies and scans of PCRs, coded under the “opioid overdose” category, were reviewed to extract the following:

        • number of PCRs recording OD,
        • clinical presentation,
        • clinical care provided (pre-post arrival),
        • number and percentage of patients receiving naloxone,
        • response to naloxone,
        • number of patients transported/refused transport, and
        • deaths, if confirmed by the ambulance service on scene.

Accuracy of data entry was assessed by an external review of a 20% random sample of records.

No follow-up clinical data for care in the ED were collected for this study. Therefore, the clinical outcomes of the overdose events cannot be commented on, other than in respect of the small number of cases in which patients were confirmed to be dead by the ambulance service, using the criteria established by the relevant Clinical Practice Guideline.

Geographical information

In 2012, the Central Statistics Office (CSO) published Small Area Population Statistics for the 2006 and 2011 censuses. Small areas (SAs) are national boundaries created by Ordinance Survey Ireland as a subdivision of preexisting electoral districts and are available for download from the CSO’s Web site (www.cso.ie/census). As the smallest Geographical areas for which census data are available, they provide the most accurate level for measurement. Small areas are standardized in size, with a minimum of 65 households and a mean of less than 100, thus effectively providing street-level information on the Irish population. Population statistics are now available for 18488 SAs in Ireland.

Deprivation index

The Pobal-Haase-Pratschke Deprivation Index is a composite score “measuring the relative affluence or disadvantage of a particular geographical area” using data compiled from 2006 and 2011 census [26]. Each area is scored from approximately -40 (being the most disadvantaged) to +40 (being the most affluent), with zero as the average national score from the 2006 census. Fourteen indicators in 3 dimensions of affluence/disadvantage comprise the deprivation index: demographic profile, social class composition, and labor market situation. For the purpose of this study, we categorized overdoses by their location into street, residential, or service overdoses. The residential category comprised house and hotel. The services included hostel/homeless shelter, treatment center, hospital, shop, bar/pub, or police station.

Methadone clinics

As of October 2010, there were 9285 patients attending metha- done treatment programs nationally, of which 3312 patients (36%) received care through general practitioners, 5368 (57.5%) in the 66 Health Service Executive clinics, and 604 patients (6.5%) were attending treatment in prison [27]. Fifty-three clinics were located in Dublin, serving 4783 patients. We added clinic locations to our geographical map. We compared presence of a clinic in the area, patient load and number of ODs in the (a) SA, (b) 500-m radius around clinic, and (c) 1000-m radius around clinic.

Data analysis

In accordance with the primary objective of this study, descriptive analysis was carried out on the key indicators: OD calls and clinical information on each call (frequencies, correlations, t tests, and ?2 test). Information extracted from PCRs was entered into an MS Excel spreadsheet. Deidentified data were then exported into an IBM SPSS 20 database, IBM Corp. in Armonk, NY, from which the statistical

Table 1

Characteristics of overdoses (n = 469, missing data reported in each variable)

Characteristic (n = 464)

Description

Patient total

Location data missing

Street, n = 212 (45.2%)

Residential, n = 178 (38%)

Services, n = 74 (15.8%)

Opioids (n = 441)

Evidence of opioid use at

418 (89.1)a

5

197 (92.9)

155 (87.1)

61 (82.4)

scene

Other substances (n = 131)

Alcohol

29 (6.2)

0

15 (7.1)

11 (6.2)

3 (4.1)

Benzodiazepines

24 (5.1)

0

10 (4.7)

9 (5.1)

5 (6.8)

Antidepressants

10 (2.1)

0

2 (0.9)

8 (4.5)

0

Multiple

45 (9.6)

1

11 (5.2)

28 (15.7)

5 (6.8)

Yes (unspecified)

15 (3.2)

0

4 (1.9)

8 (4.5)

3 (4.1)

Stimulants

8 (1.7)

0

1 (0.5)

6 (3.4)

1 (1.4)

Clinical presentation (n =

Unresponsive

183 (39)

1

99 (46.7)

57 (32)

26 (35.1)

402)

Respiratory arrest

47 (10)

0

19 (9)

17 (9.6)

11 (14.9)

Respiratory depression

55 (11.7)

0

33 (15.6)

16 (9)

6 (8.1)

Cardiac arrest

19 (4.1)

0

1 (0.5)

13 (7.3)

5 (6.8)

Decreased SPO2

3 (0.6)

0

1 (0.5)

1 (0.6)

1 (1.4)

GCS 15

12 (2.6)

0

7 (3.3)

4 (2.2)

1 (1.4)

Aggression

1 (0.2)

0

1 (0.5)

0

0

Poisoning

12 (2.6)

0

1 (0.5)

9 (5.1)

2 (2.7)

Respiratory rate 20

1 (0.2)

0

0

0

1 (1.4)

Overdose

69 (14.7)

2

11 (5.2)

41 (23)

15 (20.3)

Mean GCS scores

Pre (n = 315)

7.2

6.36

7.78

7.7

Post (n = 288)

12.3

12.63

11.78

12.59

Care provided

Ventilations (n = 118)

105 (22.4)

0

46 (21.7)

42 (23.6)

17 (23)

O2 administration (n = 295)

282 (60.1)

2

140 (66)

100 (56.2)

40 (54.1)

Naloxone (n = 387)

357 (76.1)

3

182 (85.8)

117 (65.7)

55 (74.3)

Mean dose of naloxone

0.61

.58

.63

.7

(n = 359)

Route (n = 354)

IM

308 (65.7)

3

173 (81.6)

90 (50.6)

42 (56.8)

IM/Sq

1 (0.2)

0

1 (0.6)

0

IO

7 (1.5)

0

5 (2.8)

2 (2.7)

IV/IM

38 (8.1)

6 (2.8)

23 (12.9)

9 (12.2)

First aid before ambulance

42 (9)

0

9 (4.2)

17 (9.6)

16 (21.6)

(n = 406)

Transport (n = 457)

415 (88.5)

4

189 (89.2)

154 (86.5)

68 (91.9)

Death

On the scene

13 (2.8)

0

1 (0.5)

9 (5.1)

3 (4.1)

Repeat OD (n = 358)

Confirmed

96 (26.8)

0

56 (26.4)

26 (14.6)

14 (18.9)

Mean no. of repeat ODs

2.24

2.26

2.19

2.29

Abbreviations: Sq, subcutaneous; IO, intraosseous; IV, intravenous; SPO2, oxygen saturation.

a Numbers in the “Values” column include cases with unknown location, that is, n = 469. Therefore, numbers in the location columns do not add up.

analyses were performed. Small area geographic data were down- loaded from the CSO (www.cso.ie/census). Full deprivation index data for SAs were provided by the researchers. Addresses of overdose incidents were given geographic coordinates using a free online geocoding tool (www.gpsvisualizer.com/geocoder) and mapped into SAs using geographic information system software, ArcGIS version 10.1, Esri, 380 New York Street, Redlands, CA. The geodata and data on repeat overdoses by individuals were available only for the DFB data. Ethical approval for the study was provided by Human Research Ethics Committee at University College Dublin.

Results

Demography and clinical presentation

In the 12-month study period, the ambulance services attended to 469 (DFB, 358; NAS, 111) ODs. Mean age was 33 years (range, 2-701); 80% were male. Other substances were noted in 131 cases (27.9%), and the most frequent were alcohol (29), benzodiazepines (24), antide- pressants (10), and stimulants (8); 45 cases (9.6%) noted multiple substances. The most common clinical presentation was “unrespon- sive” (39%), followed by “reported overdose” (14.7%), and respiratory depression (11.7%). The unresponsive category also included de- creased Glasgow Coma Scale score and collapse. See Table 1 for more details.

1 Three children aged 2, 4, and 4 years accidentally overdosed with parents’ methadone or pain medication.

Treatment

Naloxone was administered in 357 cases (76%), mean dose of

0.61 mg and mean number of doses of 1.5 (range, 1-5); 42.6% received 1 dose. The overwhelming majority of administrations were intramuscular with small proportions (9.8%) of IM/subcutaneous, IM/intravenous, and intraosseous administrations. Mean GCS score before the incident was 7.2 (3-15), and this increased to 12.3 (3-15) after the incident. First aid before ambulance arrival was provided in 12 instances (9%). Assisted ventilations (23%) and oxygen (60%) were administered regularly.

Outcomes

Most cases were transported to hospital (88.5%). The GCS scores did not improve for 70 people (14.9%), and 45(9.6%) had GCS post Naloxone administration lower than 8. The complete prenaloxone and postnaloxone scores were recorded in 182 PCRs (38.8%). Thirteen people were found dead, or resuscitation was discontinued on the scene during the study period. Of the 358 DFB records, there were 96 repeat overdoses (26.8%) among 36 participants–5 females (14.6%) and 31 (85.4%) males. Mean number of repeat overdoses was 2.2 (range, 1-9). Three repeat ODs were fatal.

Location, time, and incidence of overdoses

The 3 most frequent places where OD occurred were street (n = 212, 45.2%), residential (n = 178, 38%), and hostel/treatment center

Table 2

Urban overdose hotspots in Dublin City

No. of SAs (n = 176)

No. of OD (n = 353a)

No. of OD x no. of SAs

1

32

32 (9.1%)

2

13

26 (7.4%)

1

11

22 (6.2%)

2

9

18 (5.1%)

2

8

16 (4.5%)

3

6

18 (5.1%)

2

5

10 (2.8%)

6

4

24 (6.8%)

9

3

27 (7.7%)

148

1-2

160 (45.3%)

a National Ambulance

Service geodata were unavailable.

(n = 74, 15.8%). Most of the ODs (70%) took place during the day, between 12 PM and 12 AM. Based on the total number of all incidents at DFB in 2011, the overdose prevalence in the DFB data was 0.49% (358/72000 x 100). Based on the same total, the overdose incidence was 4.9 cases per 1000 cases per year.

City center vs suburbs

A total of 176 SAs with 1 or more ODs were identified in the DFB data2; 148 SAs (84%) had only 1 or 2 overdoses; however, 55% of all ODs occurred in just 28 SAs (16%). This suggests that OD was concentrated in certain SAs–28 hotspots–of Dublin city. Of those, 1 had 32 ODs. See Table 2 for more details.

The majority of ODs (86%) were concentrated in the city center; the rest were in South Dublin (6%), Fingal (5%), Dun Laoghaire (1%), or at unknown locations (2%). Figure shows DFB overdoses on the map of Dublin.

We examined 355 (99.2%) of 358 DFB ODs where data were available to compare the geographical location with the setting in which the incident occurred (residential [109], street [195], or services [51]). Street ODs were concentrated in city center/quays, whereas residential ODs were dispersed throughout Dublin regions, with smaller numbers of ODs per SA than street ODs.

Street OD SAs were more likely to be in the city center (60) than suburbs (18), compared with residential overdose SAs in the city (31) and suburbs (63); see Table 3 for details. Our ?2 test revealed that the number of residential OD SAs significantly differed by location (?2(1) = 33.04; P b .001).

Urban deprivation and methadone clinics

We explored the relationship between overdose and deprivation. The mean deprivation score for Dublin is currently negative (-5.73), not zero. Our 176 SAs with overdoses had a better average deprivation score than the city average (-4.73) indicating they were more affluent overall (range, -31 to 28). There was a small but statistically significant relationship between the number of overdoses in the area and area affluence (r = 0.15; P b .05).

However, when we compared the deprivation of overdoses that occurred in a residence/house (n = 109) with those in the street (n = 195), on average, residential ODs had a lower deprivation score (-8.8, SE = 1.33), than street ODs (-1, SE = 1.44). This difference was significant (t170 = 3.99; P b .001). For city center, the residential ODs’ deprivation score was closer to the city average.

Finally, we also explored the relationship between overdose and location of methadone clinics. Most ODs happened in the 1000-m- radius region around certain methadone clinics. This suggests that ODs do occur near methadone clinics, but not all clinics.

2 National Ambulance Service geodata were unavailable; therefore, the figures presented here are for DFB data only.

Discussion

Dublin ambulance services attend to an OD almost every day. This study established the incidence of ODs in Dublin ambulance service and found most of them occurring during the day, on the streets.

Our overdose and Death rates were comparable with other countries. The 10-year death rate of the Edinburgh addiction cohort recruited from primary care was 21% [28]. A naturalistic longitudinal study of injecting drug users in Inner city Dublin found 63% dead after

25 years [29] but principally because of blood borne infections. Luxembourg had 340 opiate- and cocaine-related fatal overdoses between 1985 and 2011 (approximately 13 annually) [30]. Budapest

had 299 fatal ODs between 1994 and 2012 (approximately 16 annually) [31]. In Scotland, the greatest risk of drug-related death was in the first 2 weeks of treatment [11]. In Kansas, an opioid mortality study detected fentanyl, methadone, or oxycodone in 789 overdoses from 2001 to 2011 (approximately 79 annually) [32].

The wider international trends show a decrease in drug-related deaths across Europe, reflected in the national figures from Ireland. The annual number of drug-related deaths and deaths among drug users decreased from 652 in 2009 to 607 in 2011 [6]. The number of heroin deaths decreased the most, from 115 in 2009 to 60 in 2011 [6]. Methadone deaths increased to 113, compared with 60 in 2010 [2]. This trend is similar in Scotland and could be due to the changes on the drug markets and increased polysubstance deaths (59% of all poisonings) [6].

The relationship shown between area affluence and overdoses is particularly significant and probably relates to the clear distinction between street and home events that we describe. Half of all events occurred in the street, and the large majority of these were quite close to drug treatment centers within the city center, which is rated as relatively affluent, giving an overall score, which appears slightly more affluent than the average for Dublin city. However, the deprivation scores associated with overdoses in homes is significantly worse than the overall figure. This strongly suggests that the homes, or current residences, of opioid users are in deprived areas, a finding that is consistent with most other data on illicit opioid use in Ireland [6].

Previous studies also showed that deprived areas may have more overdoses [16-18]. The concentration of heroin overdose in hotspots around drug treatment clinics may be attributable to a number of reasons. Addiction clinic areas have a predictable supply of metha- done, which may leak into the black market and community [16,33]. Many methadone patients continue to use illicit opioids and other drugs [7], and dealers may, therefore, be attracted to clinic areas, envisioning higher profits from drug sales to methadone patients and fuelling up the opioid load in the area [19,20]. Finally, the normalization theory suggests that social norms, related to public use of drugs, may be more lax in the clinic areas where patients are witnessed collecting their methadone every day [34].

The clinical findings are striking. In 418 (89%) of these calls, evidence of opiate use was observed, and in 357 (76%), naloxone was administered. For the 315 patients for whom a GCS score was recorded on arrival, the initial score was 7.2, with an average score of

6.4 in events occurring in the street. These GCS scores indicate very low levels of consciousness and are low enough to lead to an inability to safeguard the airway. The improvement to a mean level of 12.3 on scene, following treatment by ambulance staff, is striking evidence of the efficacy of the naloxone treatment provided. In 120 (25.6%) of cases, the patient was in respiratory or cardiac arrest or had respiratory depression requiring ventilatory support. The challenging environments in which patients received care are difficult ones in which to provide complex care, but given the severity of the presentations, it is striking that successful resuscitations were achieved frequently.

Naloxone is used to temporarily achieve complete or partial reversal of opioid effects, including respiratory depression, caused by natural or Synthetic opioids and specific analgesics [35-37]. It is a

*NAS geo-data were unavailable.

Figure. Spatial distribution of overdoses (n = 358). National Ambulance Service geodata were unavailable.

licensed medication (specific opiate antagonist) with no agonist properties and no potential for abuse and is inexpensive. Overdose education and naloxone distribution (OEND) save lives [9]. Large scale comparative studies have shown that OD death rates were reduced in cities and areas where OEND were implemented [38-40].

Despite this evidence, OEND do not exist in many countries, including Ireland. Implementation of OEND programs is hindered by policy and legal and personal barriers. For example, medical professionals are reluctant to deal with drugs issues [41]; minimal awareness of naloxone programs persists [42]. Reluctance and unawareness can be addressed in OEND training for medical and lay staff in drug agencies as well as other overdose witnesses.

Literature from other countries shows that bystanders, peers, or family members of overdose victims are most often the initial emergency responders and are best positioned to intervene immedi- ately, when the first overdose symptoms appear [9,43]. The response of these lay persons can save lives, if they are provided with naloxone. This study describes the context in which most overdoses occur in Dublin and highlights the potential for family members or other drugs users to offer prompt care, using naloxone. Recently, a Dublin advisory committee on lay naloxone administration has been established; data from this study will inform the provision of OEND programs to family members and other potential rescuers.

Study limitations

This study was limited in several ways. We were confined by the data sources, which covered only 1 city. This data were not triangulated with other sources, for example, death registry or hospitals. Caution should be used when interpreting our death rates; they are only the deaths confirmed on the scene of the incident.

Table 3

Distribution of overdoses in the city center vs suburbs (n = 172)

n = 172

Location

Total

Street

Residential

City center/suburb

Center

60 (76.9%)

31 (33.0%)

91 (52.9%)

Suburb?

18 (23.1%)

63 (67.0%)

81 (47.1%)

Total

78

94

172

* P b .001.

Patient care reports are not always reliable; they are handwritten under situational pressures and may sometimes be illegible. These data were already recorded, and we could not go back to clarify uncertainties. The NAS geodata were unavailable. Our results may underestimate the real overdose rates because most nonfatal overdoses are not reported due to a fear of criminal prosecution [44]. Although we explored the association between area deprivation, presence of addiction service, and overdose, we did not control for other area characteristics that might have influenced the overdose rates. These include, for example, characteristics of built environment, for example, dilapidation of buildings or number of dark lanes, whose anonymity or seclusion might attract drugs (ie, shooting galleries).

Despite the limitations, our study provides 1 year of data in the largest city in Ireland. Recent data estimate that more than 14000 people have used illicit opiates in this region and indicate that 85% of people seeking treatment for opiate use in Ireland do so in this region [45]. Use of smaller area units allowed more precise assessment of variables that may affect overdoses.

Conclusion

Our study found that most overdoses occur in daytime hours, near community drug service facilities. Knowing where and when overdoses happen can inform appropriate actions and allocate resources. For example, naloxone can be stored at the premises, and trained staff can administer naloxone within the hours of operation. Our findings clearly point out to a need to establish a lay-delivered overdose prevention and naloxone distribution system for OD in Ireland.

Acknowledgment

Work on this project was supported by the MERIT grant: Medical Emergency Responders: Integration and Training from Department of Health. The authors thank Pre-hospital and Emergency Care Council, Mark Dixon, Lawrence Kenna, Mary Headon, and Philip Evans for data collection and Kevin Tobin for help with geodata analysis.

References

  1. Beletsky L, Rich JD, Walley AY. Prevention of fatal opioid overdose. JAMA 2012; 308:1863-4.
  2. Health Research Board. Drug-related deaths and deaths among drug users in Ireland: 2010 figures from the National Drug-Related Deaths Index. Dublin: HRB; 2013.
  3. Health Research Board. National Drug Related Death Index. Dublin: HRB; 2012 [revised figs. 2012].
  4. Mongan D. Non-fatal overdoses and drug-related emergencies 2010. Drugnet Ireland; 2012 17-8.
  5. Tuusov J, Vals K, Tonisson M, Riikoja A, Denissov G, Vali M. Fatal poisoning in Estonia 2000-2009. Trends in illegal drug-related deaths. J Forensic Legal Med 2013;20:51-6.
  6. HRB. Drug-related deaths and deaths among drug users in Ireland: 2011 figures from the National Drug-Related Deaths Index. Dublin: Health Research Board; 2014.
  7. Cullen W, Bury G, Langton D. Experience of heroin overdose among drug users attending General practice. Br J Gen Pract 2000;50:546-9.
  8. Darke S, Zador D. Fatal heroin “overdose”: a review. Addiction 1996;91:1765-72.
  9. Bennett T, Holloway K, Bird SM. Does take-home naloxone reduce non-fatal overdose? Lancet 2014;383:124-5.
  10. Seymour A, Oliver JS, Black M. Drug-related deaths among recently released prisoners in the Strathclyde Region of Scotland. J Forensic Sci 2000;45:649-54.
  11. Cousins G, Teljeur C, Motterlini N, McCowan C, Dimitrov BD, Fahey T. Risk of drug- related mortality during periods of transition in methadone maintenance treatment: a cohort study. J Subst Abus Treat 2011;41:252-60.
  12. Cook S, Moeschler O, Michaud K, Yersin B. Acute opiate overdose: characteristics of 190 consecutive cases. Addiction 1998;93:1559-65.
  13. Cullen W, Griffin S, Bury G, Plunkett PK. Opiate overdose in Dublin. Ir J Med Sci 2003;172:46.
  14. Parkin S. Habitus and drug using environments: health, place and lived- experience. Ashgate Publishing, Ltd.; 2013
  15. Zinberg NE. Drug, set and setting: the basis for controlled intoxicant use. New Haven, CT: Yale University Press; 1986.
  16. Cerda M, Ransome Y, Keyes KM, Koenen KC, Tardiff K, Vlahov D, et al. Revisiting the role of the urban environment in substance use: the case of analgesic overdose fatalities. Am J Public Health 2013;103:2252-60.
  17. Galea S, Ahern J, Vlahov D, Coffin PO, Fuller C, Leon AC, et al. Income distribution and risk of fatal drug overdose in New York City neighborhoods. Drug Alcohol Depend 2003;70:139-48.
  18. Hembree C, Galea S, Ahern J, Tracy M, Markham T, Piper J, et al. The urban built environment and overdose mortality in New York City neighborhoods. Health Place 2005;11:147-56.
  19. City Clinic. City Clinic 20 years of reflections, remembrances and recollections 1993-2013. Dublin: Health Service Executive; 2013.
  20. Peterson NA, Lowe JB, Reid RJ. Tobacco outlet density, cigarette smoking prevalence, and demographics at the county level of analysis. Subst Use Misuse 2005;40:1627-35.
  21. Merchant RC, Schwartzapfel BL, Wolf FA, Li W, Carlson L, Rich JD. Demographic, geographic, and temporal patterns of ambulance runs for suspected opiate overdose in Rhode Island, 1997-20021. Subst Use Misuse 2006;41:1209-26.
  22. Hargreaves K, Lenton S, Phillips M, Swensen G. Potential impacts on the incidence of fatal heroin-related overdose in Western Australia: a time-series analysis. Drug Alcohol Rev 2002;21:321-7.
  23. Pre-hospital emergency care Council. Clinical Practice Guidelines. Dublin: PHECC; 2013.
  24. DFB. Clinical Audit in DFB EMS. Phecc Voice Newsletter; 2012 1-4.
  25. O’Donnell C. Ambulances aim to treat as well as transport. Forum. Dublin: ICGP; 2013 12-3.
  26. Haase T, Pratschke J, Gleeson J. All-island deprivation index: towards the development of consistent deprivation for the island of Ireland. http:// trutzhaase.eu/wp/wp-content/uploads/R-2012-All-Island-Deprivation-Index.pdf.
  27. Doyle J. Letter to Dail Eireann about the number of methadone and stabilisation centres in each drug task force area. PQ No: 39293/10; 2010 [Dublin].
  28. Macleod J, Copeland L, Hickman M, McKenzie J, Kimber J, De Angelis D, et al. The Edinburgh Addiction Cohort: recruitment and follow-up of a primary care based sample of Injection drug users and non drug-injecting controls. BMC Public Health 2010;10:101.
  29. O’Kelly FD, O’Kelly CM. The natural history of injecting drug use: a 25-year longitudinal study of a cohort of injecting drug users in inner city Dublin. Ir J Med Sci 2012;181(4):541-8.
  30. Origer A, Lopes da Costa S, Baumann M. Opiate- and cocaine-related fatal overdoses in Luxembourg from 1985 to 2011: a study on gender differences. Eur Addict Res 2013;20:87-93.
  31. Horvath M, Dunay G, Csonka R, Keller E. Deadly heroin or the death of heroin– overdoses caused by illicit drugs of abuse in Budapest, Hungary between 1994 and 2012. Neuropsychopharmacologia Hungarica: a Magyar Pszichofarmakologiai Egyesulet lapja = official journal of the Hungarian Association of. Psychopharma- cology 2013;15:253-9.
  32. Okic M, Cnossen L, Crifasi JA, Long C, Mitchell EK. Opioid overdose mortality in Kansas, 2001-2011: toxicologic evaluation of intent. J Anal Toxicol 2013;37: 629-35.
  33. Mayer JD, Kirlin B, Rehm CD, Loeser JD. Opioid availability in outpatient pharmacies in Washington state. Clin J Pain 2008;24:120-3.
  34. Murray E, Treweek S, Pope C, MacFarlane A, Ballini L, Dowrick C, et al. Normalisation process theory: a framework for developing, evaluating and implementing complex interventions. BMC Med 2010;8:63.
  35. Sporer KA, Kral AH. Prescription naloxone: a novel approach to heroin overdose prevention. Ann Emerg Med 2007;49:172-7.
  36. McDermott C, Collins NC. Prehospital medication administration: a randomised study comparing intranasal and intravenous routes. Emerg Med Int 2012;2012: 476161.
  37. Merlin MA, Saybolt M, Kapitanyan R, Alter SM, Jeges J, Liu J, et al. Intranasal naloxone delivery is an alternative to intravenous naloxone for opioid overdoses. Am J Emerg Med 2010;28:296-303.
  38. Walley AY, Xuan Z, Hackman HH, Quinn E, Doe-Simkins M, Sorensen-Alawad A, et al. Opioid overdose rates and implementation of overdose education and nasal naloxone distribution in Massachusetts: interrupted Time series analysis. BMJ 2013:346.
  39. CDC. Community-based opioid overdose prevention programs providing naloxone– United States, 2010. MMWR Morb Mortal Wkly Rep 2012;61:101-5.
  40. Coffin PO, Sullivan SD. Cost-effectiveness of distributing naloxone to heroin users for lay overdose reversal. Ann Intern Med 2013;158:1-9.
  41. BMA Board of Science. Drugs of dependence: the role of medical professionals. Oxford: British Medical Association; 2013.
  42. Pflanz-Sinclair C, Matheson L, Aucott L. Naloxone GP Research Project Steering Group. General Practitioner Engagement with the Scottish National Naloxone Programme: a needs assessment project. NHS Health Scotland: Aberdeen; 2013.
  43. Kim D, Irwin KS, Khoshnood K. Expanded access to naloxone: options for critical response to the epidemic of opioid overdose mortality. Am J Public Health 2009;99:402-7.
  44. Gaston RL, Best D, Manning V, Day E. Can we prevent drug related deaths by training opioid users to recognise and manage overdoses? Harm Reduct J 2009;6:26.
  45. O’Brien M, Moran R, Kelleher T, Cahill P. National Drug Treatment Reporting System, Statistical Bulletin 1997 and 1998. National Data and Data by Health Board Area. Dublin: HRB; 2000.

Leave a Reply

Your email address will not be published. Required fields are marked *