Social influences on the transition to injection drug use among young heroin sniffers: a qualitative analysis

1. Introduction 

Recently, a growing body of literature has focused on the period directly following initiation of injection, as this period has been found to be characterized by a high risk of becoming infected with HIV and hepatitis C virus (HCV) (Nicolosi, Leite, Musicco, Molinari & Lazzarin, 1992, van Ameijden, Van Den Hoek, Hartgers & Coutinho, 1994, Carneiro, Fuller, Doherty & Vlahov, 2000). As the first injection may not be planned, new IDUs might engage in the risky behavior or being injected by someone else because of lack of experience, the fear of injecting, or the fact that they do not usually have their own injection equipment (Kral, Bluthenthal, Erringer, Lorvick & Edlin, 1999). These factors could lead to injecting with unclean syringes. Additionally, new injectors may not be aware of all of the risks associated with sharing used injection equipment.

Although studies have focused on preventing the initiation of injection, a growing body of quantitative research has focused on the factors associated with the transition from non-injection to injection drug use (Crofts, Louie, Rosenthal & Jolley, 1996, Hunt, Griffiths, Southwell, Stillwell & Strang, 1999). Among Dutch drug users, van Ameijden and colleagues (1994) found that the 5-year cumulative incidence of transitioning to injection was 30%. Significant risk factors for transitioning to injection drug use have included: having an injecting sex partner (Vlahov et al., 1991), having a lower level of schooling (Dunn & Laranjeira, 1999; Crofts et al., 1996), having a history of sniffing or smoking heroin or cocaine (Fullilove, Fullive, Haynes & Gross, 1990, Dunn & Laranjeira, 1999), and reporting a high level of addiction to heroin and or cocaine (Crofts et al., 1996; Matthias, 1991). Few studies have explicitly focused on the factors influencing transition to injection drug use among young heroin sniffers before they initiate injection drug use.

As early as 1989, Des Jarlais and colleagues reported that among heroin sniffers, transition into injection drug use seemed to be more likely among those whose friends or relatives injected drugs. However, studies considering the role of the social environment on transition behaviors are sparse. Research examining the social, economic, and interpersonal factors that influence an individual's decision to transition to injecting drugs is vital in the design of effective HIV prevention programs targeting young drug users and identifying treatment modalities that best meet the needs of young drug users.

To date, a theoretically-based understanding of the social dynamics surrounding the event of transitions from non-injection to injection is lacking in the literature. Social influence theory is a useful framework in understanding how the social environment directly and indirectly affects individuals’ behavioral choices (Fisher & Fisher, 1992, Latkin, 1998). Social influence usually occurs through social norms, behavioral beliefs that condone and condemn specific behaviors which have a potent effect on members’ attitudes and behaviors (Fisher & Misovich, 1990). In one of the only studies examining the event of first injection among 300 young IDUs, Crofts et al. found that 88% were injected for the first time by a friend. Studies of drug users’ social networks have demonstrated the influence of peer networks on individuals’ health promotion and risk taking behaviors. A growing body of literature utilizes social networks as a means to promote lasting behavior change among groups of IDUs at risk for HIV (Trotter, Bowen & Potter, 1995, Latkin, Mandell, Vlahov, Oziemkowska & Celentano, 1996, Broadhead et al., 1998). Among IDUs, perceived normative expectations and peer pressures are determinants of risky injection behaviors (Latkin et al., 1996).

The aim of this qualitative study was, through the framework of social influence theory, to explore the spheres that influence young drug users’ transition from heroin sniffing to injecting in Baltimore, MD, USA, where it is estimated that the number of persons requiring treatment for heroin addiction is 15 times that of the US national rate (Smart Steps). In an effort to more fully understand the factors that led up to the point of injection, we also examine the spheres of influence on initiation and development of habitual heroin use. The qualitative study was embedded in a larger longitudinal quantitative study of injection and noninjection drug users. The smaller qualitative study was conducted in an effort to obtain a deeper understanding of the environmental factors that influence transition to injection among young drug users.

2. Methods 

The study sample was selected from participants in a larger study of young IDUs, the Risk Evaluation and Assessment of Community Health III (REACH III) study. The purpose of REACH is to identify the prevalence and correlates of blood borne infections among recent initiates to injection. Persons eligible for REACH are those who initiated injection drug use within the prior 5 years, reside in the Baltimore metropolitan area, and are between the ages of 15 and 30 years. At baseline and semi-annually, REACH participants undergo an interviewer-administered questionnaire and venipuncture for testing of HIV and HCV antibody. Upon request, participants are also provided referrals to available services, such as the needle exchange program, drug treatment and medical care.

Since the present study aimed to explore circumstances prior to the initiation of injection, we restricted the study sample for this investigation to new initiates, as defined by those who had first injected with the previous 3 years. The sample was a convenience sample (n=19), with an oversampling of women to ensure representation.

Data were collected using an open-ended interview guide which asked about: individual's daily routine; family history; experiences using drugs for the first time; experiences using heroin for the first time; and the events and individuals surrounding their first time injecting. The guide was developed through extensive discussion with young drug users as well as the broader study staff to incorporate the appropriate domains of inquiry. After the guide was initially developed, it was pilot tested among four young, new initiates to injection and finalized after this process. Informed consent was obtained from all participants prior to interview.

Qualitative methods allowed participants to describe, in their own words, the meaning of their first injection experience embedded in the broader context of their personal drug use history and broader community. Data collection occurred on a mobile van that had been modified to include a private room for confidential interviews. All interviews were tape-recorded and were administered by two trained interviewers (first two authors). Interviews were transcribed from audiotapes and transcripts were cross-validated against the tapes to ensure accuracy. Additionally, for each participant, sociodemographic and behavioral data were obtained from the parent study. Both the present and the parent REACH study were approved by the Johns Hopkins University Committee on Human Research.

All transcripts were coded for common themes that emerged across the data. To ensure coding reliability and validity, both interviewers developed separate coding schemes based on 25% of the transcripts. Once a coding scheme was agreed upon and validated with key informants, hard copies of the transcripts were coded. Each coded transcript was then entered into the qualitative analysis program NUD*IST, which allowed for a complex analysis of themes by combining all of the coded transcripts into one data set. At this point, the data was reconstituted in a new form, tied together by the themes that developed across as opposed to within interviews. The codes discussed in this analysis are those that relate to the influences, both at the social and individual levels, that participants identified as influencing their transition to injection drug use.

3. Results 

The spheres of influence that emerged from the data occurred on both the social/contextual and individual levels and included: family; sex partners/spouses; friends; and the physical environment of the neighborhood. Individual influences included the level of addiction and maintenance of habit as well as the expense of sniffing. All names used in this analysis are pseudonyms.

Study participants (n=19) were 55% male, 95% Caucasian, and 61% had not completed high school. The mean age was 24 years of age (range: 19–29 years old) and 61% reported being unemployed. None tested positive for HIV antibodies at the time of their baseline interviews for the larger REACH study, but ten tested positive for HCV antibodies. Participants reported first snorting heroin and/or cocaine at an average age of 17 years old and reported first injecting heroin at an average age of 21.7 years. The median time to injection from sniffing was 4 years. Ninety percent reported injecting heroin daily, of whom 61% injected over five times a day. When asked with whom they injected heroin for the first time, 44% reported injecting with a friend, 34% reported injecting with a sexual partner, 17% reported injecting with a family member, and 5% reported injecting alone.

4. Discussion 

Results from this small qualitative study indicate that individuals’ initiation of drug injection followed an increasing role of drugs in the lives of these young users. This trajectory was socially sanctioned and sometimes initiated by family members, and was condoned and most often initiated with friends. The confluence of these influences normalized the use of various drugs and influenced participants’ injection initiation.

Social influence theory is a useful framework in examining the factors associated with transition to injection drug use. Social influence occurs through an individual's proximal social environment, or their social network, by the establishment of social norms. Social norms are group characteristics that provide the legitimacy of others to influence behaviors. Social norms play a central role in the maintenance of social behaviors and present barriers to altering social behaviors (Myers & Bishop, 1970, Newcomb, 1958, Tittle, 1977, Coleman, 1990, Bettenhausen & Murrigan, 1991). Research suggests associations between social norms and HIV risk and protective behaviors (Latkin, 1995, Broadhead et al., 1998). Less attention has been placed on the processes of norm formation and change and the specificity of norms (Rhodes, Stimson & Quirk, 1996). Results from this study indicate the range of social arenas in which drug use was considered normative behavior and, therefore, easily adopted by study participants. Future research should examine the specific roles that social norms play in transition. Once a better understanding is achieved, social norms can be a focus of interventions targeting sniffers not to transition to injection.

This study supports previous research that documents the assertion that drugs users are generally introduced to injection by a friend, sex partner, or a relative (Stenbacka, 1990). In addition to parents who did not use drugs and actively discouraged participants’ drug use, many participants’ families normalized the use of drugs as a viable coping mechanism in reaction to stressful situations, in spite of the fact that many of their parents’ expressed remorse in exposing them to drug and alcohol use. Perhaps channeling parents’ feelings of guilt into positive avenues such as family-based treatment centers could help to break the intergenerational cycle of substance use. Family-based substance use treatment, which involves the family in an individuals’ recovery process, is a more holistic in their approach of treating addiction a broader social context to promote sustainability of sobriety (Liddle, 1999, Ozechowski & Liddle, 2000). Such programs have demonstrated effectiveness in working with adolescent substance users and could be used with adult users to break the cycle of inter-generational substance abuse that characterized the lives of these young IDUs.

Many of the study's participants discussed transitioning to injection as an attempt to reduce the amount of heroin they consume daily. In the eyes of the participants, injecting was a cost effective method to curtail their spending on heroin. Unfortunately this method backfired, in that the amount of heroin consumed on a daily basis quickly escalated to that when they were snorting heroin. Prevention programs should include such messages, based on the experiences of IDUs, which might dispel some of the myths of the benefits of injecting heroin to those drug users who have not transitioned.

The current study demonstrates the importance of not only examining the influence of social networks on young and newly initiated IDUs’ behaviors but to utilize social networks in interventions targeting IDUs and heroin sniffers. The influence of family and friendship networks on participants’ drug use was evident and could be incorporated into effective harm reduction interventions.

Sexual partners were a primary factor in injection initiating for women, as has been found in previous research (Ouellet, Rahimian & Wiebel, 1998). In examining IDUs’ social relationships as related to risk behavior, Rhodes & Quirk (1998) characterized sexual partnerships as a form of risk management. Through qualitative interviews, they found that some IDUs preferred having drug-using partners because of the ease of sharing the daily responsibility of copping drugs and natural understanding of the drug use lifestyle. Although these relationships rendered it difficult to stop using drugs and influenced the normalization process of use, they were also seen as providing them mutual support and protecting them from other hazards in their life. Interventions targeting IDUs with their non-injecting sexual partners are needed to intercept this dynamic.

These data are subject to the limitation of a small sample size. Although consistent themes were generated, a larger study would serve to validate this study's findings. More intense ethnography such as long-term participant observation would have served to validate and expand the study's findings. The study is also limited in that participants were recruited from one neighborhood, where there are predominantly White injection drug users. Although the ethnicity of this study's participants reflect that of the larger REACH study, ability to generalize results is limited to White, new, IDUs in Baltimore.

As this study has shown, injection drug use is a learned behavior and is related to individuals’ patterns of social relationships. Therefore, researchers and interventionists are challenged to address injection drug use as a social phenomenon in their conceptualization of research questions, areas of interest studied, and interventions that are designed to combat the issue of injection drug use. Interventions with the goal of preventing transition to injection drug use among young drug users, must start ideally before a person begins injecting and target the variety of factors that influence injection initiation and support an injecting career. In this study, the median time to injection from sniffing was 4 years, presenting a window of opportunity to intervene prior to injection initiation.