Purchase and Use Patterns of Heroin Users at an Inner-city Emergency Department
Article Outline
Abstract
Background
Many consider heroin abuse a problem of the inner city, but suburban patients may also be at risk.
Objective
To characterize the demographics and purchase/use patterns of heroin users in an inner-city emergency department (ED).
Methods
The study was conducted in one of the most impoverished and crime-ridden cities in the United States. Demographics and substance use habits of ED patients were prospectively collected. Patients who were
<18 years of age, cognitively impaired, or did not speak English were excluded. Participants were further categorized as homeless, inner-city, and suburban residents.
Results
Of 3947 participants, 608 (15%) used an illicit substance in the past year, with marijuana (9%) and cocaine (6%) the most commonly used. Heroin ranked third, used by 180 (5%) participants, with 61% male, 31% black, and 20% Hispanic. There were 64 homeless, 60 suburban, and 56 inner-city heroin users. The most common route of use was injection (68%), with the highest rate in the homeless (84%). The majority of homeless and inner-city users bought (73%, both groups) and used (homeless 74%, inner city 88%) in the inner city. Of suburban users, 58% purchased and 61% used heroin in the inner city. Prescription narcotic use was more common in homeless (20%) and suburban (23%) heroin users than in inner-city users (9%) (p<0.001).
Conclusions
Heroin is the third most commonly used illicit substance by ED patients, and a significant amount of inner-city purchase and use activity is conducted by suburban heroin users.
Keywords: emergency medicine, heroin, homeless persons
Introduction
In 2008, 2 million of the 119 million emergency department (ED) visits in the United States were associated with drug use. Of these visits, over 200,000 were related to heroin, and another 90,000 were due to the misuse of prescription narcotics (1). These findings are alarming, given that they represent an increasing trend in illicit substance abuse rates dating from 2005 (1). More concerning is that these data likely represent an underestimate of the true prevalence of illicit drug use in the ED population, because only the visits directly related to patient substance abuse are noted.
An additional concern is that illicit drug use seems to be spreading from inner cities to outlying suburban areas (2). For decades, illicit drug use has been limited to inner-city populations by geographic and socioeconomic factors. Inner cities serve as natural hubs of drug activity linking well-established drug trafficking routes with busy metropolitan airports (2). Higher rates of illicit drug use by inner-city residents are theorized to be a byproduct of poor socioeconomic conditions, including low levels of education, high rates of unemployment, and high crime rates 2, 3, 4. Over the past decade, however, heroin use in suburban populations has increased dramatically 5, 6, 7, 8. One explanation for this spread is that increased heroin production and purity has encouraged intranasal use by suburban users (5). Other evidence suggests that suburban prescription narcotic abusers may be switching over to heroin, given its more rapid onset of action and its relatively lower cost and greater ease of acquisition when compared to prescription narcotics 6, 7.
Goals of this Investigation
We attempted to describe the distribution of illicit substance use in three patient populations: the homeless, those who live in the inner city, and those who live in outlying suburban areas. Using these three categorizations, we attempted to further delineate heroin usage, including age of initiation, route of use, amounts, associated costs, and place of purchase and use. We additionally investigated socioeconomic consequences of heroin use, including changes in employment status, changes in place of residence, and prior arrests.
Materials and Methods
Study Design
We conducted a prospective, cross-sectional, observational study of a convenience sample of ED patients.
Setting
The study was conducted in an inner-city academic ED with an annual census of approximately 51,000. The ED is located in Camden, NJ, one of the most impoverished cities in the country, with 35.5% poverty rate and an average per capita income under $10,000 (9). Camden also consistently ranks as the most dangerous city in the United States, with murder and aggravated assault rates that are ranked third in the country (10). Camden is somewhat unique in that the city is well-circumscribed geographically, bordered on the west and north by the Delaware and Cooper Rivers, respectively. In the areas that are adjacent to other townships, the transition from inner city to suburbs typically occurs within the span of one to two city blocks.
The study was conducted over 4 months in 2007. Written informed consent was obtained by all participants and the study was approved by our institutional review board.
Selection of Participants
Trained research associates screened patients from 9:00 a.m. to 10:00p.m. 7 days a week in the adult section of the ED. Patients aged 18 years or older were approached for inclusion and given a verbal description of the study before written informed consent was obtained. Patients who were cognitively impaired, critically ill, or who did not speak English were excluded. Those who seemed intoxicated were reassessed by medical staff. After patients were deemed no longer intoxicated as per the treating physician, they were referred to the research assistants for enrollment in the study. Potential participants were obtained from the general ED population; the patient’s visit did not need to be related to substance use or misuse.
Data Collection and Processing
The questionnaire included demographic items, level of education, and employment status. Triage status, as a measure of patient acuity, was also recorded. Each participant was asked to note their drug use. For those who responded positively to heroin use within the past year, additional information about socioeconomic status as well as prior and current heroin purchase and use habits were obtained.
Participants were further subcategorized as homeless, inner-city residents, and suburban residents. Homeless patients were self-identified; no formal criterion for homelessness was applied. Both zip code and city of residence were obtained from those who had a domicile and were cross-referenced with each other for accuracy. For Camden, there are five zip codes: 08101–08105. Camden city is unique in that the areas within these five zip codes correlate to the geographic boundaries of the “inner city.” Thus, we were able to corroborate inner-city resident status using both self-reported city of residence and zip code.
Primary Data Analysis
Descriptive statistics were used to describe sample characteristics. Continuous variables are presented as means±SD, and for non-parametric data, medians and interquartile ranges are noted. Chi-squared and one-way analysis of variance were used to compare the three groups; data were analyzed using SPSS version 16.0 (SPSS, Inc., Chicago, IL).
Results
During the 4-month period, 6627 patients were screened and 3947 (60%) were enrolled. Potential participants were excluded for the following reasons: patient was critically ill or in acute distress and unable to complete the questionnaire, 1061 (16%); refused, 692 (10%); cognitive impairment that did not improve during the ED visit, 493 (7%); did not speak English, 280 (4%); and had completed survey in prior visit, 154 (2%). The remaining 3947 patients completed questionnaires and provided informed consent. Participant demographics are presented in Table 1.
Table 1. Participant Characteristics
| Characteristic | Homeless (n=339) | Inner City (n=1606) | Suburban (n=2002) |
|---|---|---|---|
| Age (years), mean (SD) | 44.4 (13.9) | 43.3 (16.7) | 47.4 (18.5) |
| Male sex (%) | 230 (68) | 690 (43) | 979 (49) |
| Race | |||
| White | 172 (51) | 587 (37) | 1435 (72) |
| Black | 154 (45) | 929 (58) | 489 (24) |
| Asian | 0 | 8 (0.5) | 24 (1) |
| Other | 13 (4) | 82 (5) | 54 (3) |
| Hispanic | 59 (17) | 503 (31) | 205 (10) |
| Education: Did not graduate high school | 130 (38) | 704 (44) | 380 (19) |
| Unemployed | 184 (54) | 544 (34) | 439 (22) |
| Receiving welfare | 52 (15) | 162 (10) | 99 (5) |
| Triage level, mean | 3.2 (0.9) | 3.2 (0.8) | 3.0 (0.8) |
The greatest proportion of illicit substance use was found in the homeless population, with cocaine, heroin, and marijuana being the most frequently reported drugs. Illicit substance usage of the homeless, inner-city, and suburban subgroups is compared in Table 2.
Table 2. Participant Illicit Substance Use
| Used in the Past Year, n (%) | Homeless (n=339) | Inner City (n=1606) | Suburban (n=2002) | p Value∗ |
|---|---|---|---|---|
| Any illicit substance | 131 (39) | 244 (15) | 233 (12) | < 0.0001 |
| Marijuana | 53 (16) | 160 (10) | 154 (8) | 0.001 |
| Cocaine | 92 (27) | 75 (5) | 87 (4) | < 0.0001 |
| Heroin | 64 (19) | 56 (4) | 60 (3) | < 0.0001 |
| Prescription narcotics | 17 (5) | 9 (1) | 27 (1) | < 0.0001 |
| Sedatives | 9 (3) | 4 (< 1) | 9 (1) | |
| Phencyclidine (PCP) | 9 (3) | 10 (1) | 8 (< 1) | |
| Amphetamines | 7 (2) | 3 (< 1) | 6 (< 1) | |
| Ecstasy | 6 (2) | 7 (< 1) | 8 (< 1) | |
| Inhalants | 2 (1) | 0 | 4 (< 1) | |
| Lysergic acid diethylamide (LSD) | 2 (1) | 2 (< 1) | 4 (< 1) | |
| Cold medications | 0 | 8 (1) | 8 (< 1) | |
| Other | 3 (1) | 10 (1) | 8 (< 1) | |
| Mean number of drugs used (SD) [Range] | 2.0 (1.4) [1–7] | 1.4 (0.9) [1–8] | 1.6 (1.3) [1–9] | < 0.0001 |
∗Significance testing was conducted for only the four mostly commonly used drugs. |
Participants who reported using heroin in the past year provided additional details of their use. There were 180 self-reported heroin users in this cohort, which were fairly evenly distributed among the three groups. The age of first use did not differ among groups, nor did arrest rates. Residence changed for many participants since their first use. Less than half of homeless participants resided in Camden with first heroin use and only two-thirds of current inner-city residents lived in Camden with first heroin use. Not one of these participants relocated from another inner city; the majority moved from the local suburbs or from another suburb or metropolitan area outside New Jersey. Of participants currently living outside the inner city, nearly one-third (28%) relocated from Camden after first heroin use. Table 3 presents the characteristics of participants who used heroin in the past year and compares the homeless, inner-city, and suburban groups.
Table 3. Characteristics of Participants Who Have Used Heroin in the Past Year
| Characteristic, n (%) | Homeless (n =64) | Inner City (n =56) | Suburban (n =60) | p Value |
|---|---|---|---|---|
| First illegal drug | ||||
| Marijuana | 44 (69) | 39 (70) | 36 (60.0) | 0.47 |
| Heroin | 4 (6) | 10 (18) | 5 (8) | 0.09 |
| Age at first use heroin, mean (SD) | 22.0 (8.7) | 23.6 ( 9.7) | 23.3 (8.8) | 0.63 |
| Current age, mean (SD) | 35.6 (11.2) | 41.1 (10.9) | 34.2 (9.7) | 0.001 |
| Unemployed before using heroin | 14 (22) | 17 (30) | 12 (20) | 0.38 |
| Unemployed currently | 46 (72) | 30 (54) | 28 (47) | 0.01 |
| Arrested for heroin use∗ | 24 (41) | 22 (42) | 21 (46) | 0.90 |
| Lived in inner city with first heroin use† | 24 (43) | 32 (67) | 13 (28) | < 0.001 |
| Currently buy heroin in the inner city | 47 (73) | 41 (73) | 35 (58) | 0.13 |
| Currently use in the inner city‡ | 41 (75) | 44 (88) | 28 (61) | 0.009 |
| Used heroin in the past 30 days | 61 (95) | 55 (98) | 47 (78) | < 0.001 |
| Route of use§ | ||||
| Snort | 18 (28) | 25 (45) | 27 (45) | |
| Inject | 54 (84) | 32 (57) | 36 (60) | 0.002 |
| Smoke | 10 (16) | 6 (11) | 7 (12) | |
| Inhale | 6 (9) | 6 (11) | 3 (5) | |
| Skin pop | 5 (8) | 4 (7) | 3 (5) | |
| Amount spent on heroin (past 30 days), median (IQR) [range] | $800 ($1880) [$0–30,000] | $250 ($965) [$0–$25,000] | $300 ($1368) [$0–$10,000] | 0.03 |
| Also use prescription narcotics | 13 (20) | 5 (9) | 14 (23) | 0.10 |
| Would participate in needle exchange‖ | 48 (89) | 22 (63) | 35 (90) | 0.002 |
∗Fifty-eight homeless, 52 inner-city, and 46 suburban dwellers responded to these item. |
†Fifty-six homeless, 48 inner-city, and 46 suburban dwellers responded to this item. |
‡Fifty-five homeless, 50 inner-city, and 46 suburban dwellers responded to this item. |
§Comparisons were only made for injection heroin use. |
‖Only includes those who inject or skin pop and have used heroin in the past 30 days; 54 homeless, 35 inner-city, and 39 suburban dwellers responded to this item. |
Purchase and use patterns were also noted. The majority of inner-city and homeless residents and over half of suburban users purchased and used heroin in Camden (Table 3). The second most common purchase site was Philadelphia, PA, with six homeless, four suburban, and one inner-city user purchasing there. Philadelphia was also the second most common site of use, with four homeless, four suburban, and two inner-city participants reporting heroin use there.
Route of use was predominantly via injection, with homeless participants more likely to inject than inner-city and suburban heroin users. Snorting was the second most favored route of use of participants. Median 30-day expenditures for heroin seemed to be in disagreement with residency status, with the highest median dollar amount among homeless participants (Table 3).
Discussion
In this investigation, the greatest proportion of substance use was in the homeless population, with 39% having used an illicit substance in the past year. In comparison, only 15% of inner-city participants and 12% of suburban participants reported drug use within the past year. In keeping with established trends, marijuana was the most commonly reported first drug of abuse in all three groups (8). After marijuana, cocaine and heroin were the second and third most commonly used drugs by participants. The highest illicit substance usage rates were reported by homeless participants, which replicate findings of a previous investigation of homeless substance users in Pittsburgh and Philadelphia (11). An unexpected finding was that, overall, the abuse of prescription narcotic medications was most prevalent in the homeless population. Yet, when only heroin users were examined, high rates of prescription narcotic use were found in both the homeless and suburban residents. For the homeless, this increased use of prescription narcotics may reflect the use of local health care facilities as a means to supplement their narcotic supply. In the suburban heroin users, this finding may reflect the process of a conversion from prescription narcotic use to the less expensive, purer, and more rapidly effective heroin 6, 7.
Our data also demonstrated a shift in participant demographics. Although heroin use is primarily considered a problem of inner cities, 54% of participants lived in the suburbs at the time of first heroin use. For current heroin users, nearly 60% of homeless participants and approximately one-third of current inner-city residents originated from outside the inner city, suggesting an association between heroin use and a decline in housing status (either via a geographic shift into the inner city or the loss of one’s home). This is supported by data from O’Toole et al., where 59% of respondents cited drug or alcohol use as the major reason for their homelessness (11). In contrast, less than one-third of our suburban cohort originally lived in the inner city with first heroin use.
Interest in a needle exchange program was greatest in the homeless and suburban heroin users. Increased interest by participants, particularly the homeless, might be due to economic factors, to allow for increased access to needles 12, 13. Increased interest by suburban users (compared to inner-city heroin users) may be due to a higher level of education (and understanding of risk) in suburban participants (14).
Of self-reported heroin users, the distribution among homeless, inner-city, and suburban users did not differ. As expected, the majority of homeless and inner-city heroin users bought and used heroin within the inner city. A sizable percentage of suburban users also admitted to buying and using heroin within the inner city. Although it was expected that purchase might occur in the inner city, it was not anticipated that over 60% of suburban participants would also report using heroin in the inner city. In retrospect, our findings are reflected in our clinical experience where suburban patients present to the ED with respiratory depression after a heroin misadventure. Our findings confirmed our suspicion that the problem of heroin trafficking and abuse is not a problem limited to inner-city residents and the homeless.
Finally, the median expenditure amounts reported by participants initially seemed inconsistent with housing status, with several homeless individuals reporting very large sums. Although not formally part of the questionnaire, several homeless participants with high expenditures admitted that these costs reflected drug trafficking activity.
Limitations
There are several limitations of this investigation. This was a convenience sample, and no screening occurred overnight, the period when substance abusers more frequently present. This likely affected our drug use prevalence estimates. Another potential cause for an underestimate in drug use was that cognitive impairment and critical illness were exclusions to participation due to the requirement for informed consent. Although many of these patients were approached and enrolled after several hours of treatment and stabilization in the ED, we still were not able to capture all of them. Furthermore, due to the sensitive nature of this investigation and the vulnerable population (homeless) that was enrolled, our institutional review board did not allow us to record demographic characteristics or the specific reason for exclusion due to cognitive impairment (intoxication vs. dementia). Therefore, we were unable to determine whether or how this group of patients differed from our cohort. Concern about a breach of confidentiality or differential treatment by ED staff may also have led participants to underreport their substance abuse. We attempted to minimize these concerns by having dedicated research staff administer the questionnaires after medical evaluations were completed. Participants were further advised that their questionnaire responses would not be shared with medical staff. We also cannot predict how generalizable our findings are in relation to other urban areas, given the unique well-circumscribed geographic and demographic characteristics of Camden, NJ. We suspect that investigations in cities with relatively clearly demarcated inner-city and suburban areas within close range such as Philadelphia, Washington, DC, and Chicago would provide similar findings 15, 16. Despite these limitations and the sensitive nature of the questionnaire items, the participation rate (69% of cognitively intact and newly approached patients) was higher than anticipated. Moreover, once patients agreed to participate, not one withdrew consent and only a very small number of highly sensitive items were left unanswered by a small number of participants (Table 3). A final limitation is that we did not formally categorize participants as homeless using established criteria (e.g., using the federal definition of homelessness, US Code, title 42, chapter 119) (17). Instead, participants self-identified their living situation, which may have yielded an underestimate of the proportion of homeless participants in our sample.
Conclusions
Heroin was the third most commonly used illicit substance by ED patients, with self-reported use equally distributed among the homeless, inner-city, and suburban residents. Prior investigations have noted that rates of heroin use are increasing in the suburbs; however, our data are the first to demonstrate that a substantial amount of inner-city purchase and use activity is conducted by suburban heroin users. These findings suggest that ED-based screening and referral efforts may need to be adjusted to account for the increasing proportion of suburban users. Furthermore, there are public policy implications if suburban users are purchasing and using disproportionately in the inner city. The drug problems that plague inner cities should have a regional approach. Consideration should be given for inclusion of suburban areas for needle exchange programs and well as methadone clinics. Likewise, awareness programs should be initiated for suburban physicians to not only screen for narcotic abuse but also to carefully monitor prescription narcotic dispensation, given that prescription narcotic abuse is often the first step toward injection drug use in suburban patients.
Article Summary
1. Why is this topic important?
2. What does this study attempt to show?
3. What are the key findings?
4. How is patient care impacted?
References
- Drug Abuse Warning Network. Available at: https://dawninfo.samhsa.gov/. Accessed February 16, 2010.
- . Heroin hits the suburbs–hard. A sad trail of abuse is spreading beyond the Northeast’s inner cities. US News World Rep. 2008;145:27–28
- . Injection drug users in the United States, 1979–2002: an aging population. Arch Intern Med. 2007;167:166–173
- Unfair treatment, neighborhood effects, and mental health in the Detroit metropolitan area. J Health Soc Behav. 2000;41:314–332
- . National drug threat assessment 2009: heroin. Available at: http://www.usdoj.gov/ndic/pubs31/31379/heroin.htm#TopAccessed February 16, 2010
- . Racial/ethnic disparities in injection drug use in large US metropolitan areas. Ann Epidemiol. 2005;15:326–334
- . Racial and ethnic changes in heroin injection in the United States: implications for the HIV/AIDS epidemic. Drug Alcohol Depend. 2008;94:221–233
- . Results from the 2007 National Survey on Drug Use and Health: national findings (Office of Applied Studies, NSDUH Series H-34, DHHS Publication No. SMA 08-4343). Rockville, MD: Substance Abuse and Mental Health Services Administration; 2008;
- . US Census Bureau website. Available at: http://www.census.gov/Accessed February 16, 2010
- In: O’Leary Morgan K, Morgan S, Boba R editor. City crime rankings 2008–2009: crime in metropolitan America. Washington, DC: CQ Press; 2008;Available at: http://os.cqpress.com/citycrime2008/citycrime2008.htmAccessed February 16, 2010
- . Self-reported changes in drug and alcohol use after becoming homeless. Am J Public Health. 2004;94:830–835
- . Unstable housing as a factor for increased injection risk behavior at US syringe exchange programs. AIDS Behav. 2007;11(6 Suppl):78–84
- . Persistence and change in disparities in HIV infection among injection drug users in New York City after large-scale syringe exchange programs. Am J Public Health. 2009;99(Suppl 2):S445–S451
- Social and political factors predicting the presence of syringe exchange programs in 96 US metropolitan areas. Am J Public Health. 2007;97:437–447
- . Pennsylvania drug threat assessment, June 2001. Available at http://www.justice.gov/ndic/pubs0/670/heroin.htmAccessed April 30, 2010
- . Illinois drug threat assessment, January 2001. Available at http://www.justice.gov/ndic/pubs/652/heroin.htmAccessed April 30, 2010
- Legal Information Institute, United States Code, Title 42, chapter 119. Available at: http://www.law.cornell.edu/uscode/uscode42/usc_sup_01_42_10_119.html. Accessed August 10, 2010.
PII: S0736-4679(10)00516-0
doi:10.1016/j.jemermed.2010.06.006
© 2012 Elsevier Inc. All rights reserved.
