Reasons for Selecting an Initial Route of Heroin

7/29/2019 Reasons for Selecting an Initial Route of Heroin

1/12

RESEARCH REPORT

2003 Society for the Study of Addiction to Alcohol and Other Drugs Addiction, 98, 749760

Blackwell Science, LtdOxford, UKADDAddiction1360-0443 2003 Society for the Study of Addiction to Alcohol and Other Drugs98Original ArticleHeroinadministrationroutesduringHIVepidemicMaraJ.Bravo

etal.

Correspondence to:

Luis de la Fuente

Centro Nacional de Epidemiologa

Instituto de Salud Carlos III

Calle Sinesio Delgado 628029 Madrid

Spain

Tel: +34 91 387 75 08

E-mail: [email protected]

Submitted 8 August 2002;

initial review completed 25 October 2002;

final version accepted 3 February 2003

RESEARCH REPORT

Reasons for selecting an initial route of heroin

administration and for subsequent transitions during a

severe HIV epidemic

Mara J. Bravo1, Gregorio Barrio2, Luis de la Fuente1,3, Luis Royuela2, Laura Domingo3 &Teresa Silva3

Secretara del Plan Nacional sobre el Sida, Madrid,1 Centro Universitario de Salud Pblica (CUSP), Madrid,2 Proyecto Itnere, Centro Nacional de Epidemiologa,

Instituto de Salud Carlos III, Madrid, Spain3

ABSTRACT

Aim To identify the most important reasons for selecting a particular route of

heroin administration and for subsequent transitions during a period of epi-demic HIV transmission. To study temporal trends in these reasons.

Design Cross-sectional survey.

Participants Nine hundred heroin users in three Spanish cities: 305 in Seville,

297 in Madrid and 298 in Barcelona.

Measurements A separate analysis was made of the reasons for five types of

behaviour: (a) selecting injection as the initial usual route of heroin adminis-

tration (URHA); (b) changing the URHA to injection; (c) never having injected

drugs; (d) selecting the smoked or sniffed route as the initial URHA; and (e)

changing the URHA to a non-injected route. Subjects were invited to evaluate

the importance of each reason included in a closed list. Spontaneously self-

perceived reasons were also explored in an open-ended question for each of the

five types of behaviour studied.Findings The primary reason selected for each type of behaviour was: (a) pres-

sure of the social environment; (b) belief that injection is a more efficient route

than smoking or sniffing heroin; (c) concern about health consequences (espe-

cially fears of HIV and overdose), and fear of blood or of sticking a needle into

ones veins; (d), pressure of the social environment and (e) concern about health

consequences and vein problems. For women, having a sexual partner who

injected heroin played a decisive role in initiating or changing to injection. Few

people spontaneously mentioned market conditions for purchasing heroin as

an important reason for any behaviour, nor did many mention risk of overdose

as reasons for (c) or (d).

Conclusions These findings should be considered when designing interven-

tions aimed at preventing initiation of injecting or facilitating the transition tonon-injected routes.

KEYWORDS Administration route, heroin, injecting, smoking, sniffing,

transitions.

INTRODUCTION

The three most frequent routes of heroin administration

are injection, smoking and sniffing. The use of a particu-

lar route has important implications for a users health

and social life. The injected route generates by far the

most severe health problems, primarily infections and

overdose [13]. These problems, particularly infections


2/12


750 Mara J. Bravo et al.

such as HIV/AIDS, HCV and HBV, which are associated

with sharing contaminated injection material, are major

threats to public health in many countries.

Consequently, preventing the use of the injected route

and facilitating the change to other routes of administra-

tion are important public health objectives [4]. These are

feasible objectives as previous studies have shown that

the usual route of heroin administration (URHA) can

change during a drug users career, that the proportion of

heroin users whose URHA is injection varies greatly by

geographical area, and that this proportion has

decreased in recent years in some countries [58].

Despite this evidence, few interventions have been devel-

oped to prevent initiation of injecting or to facilitate the

change to other routes, except for methadone mainte-

nance programmes [911].

Studies have found that the main reasons for adopting

or maintaining the injected route are its superior effi-

ciency compared to the smoked or sniffed routes[1217],the fact of having used the injected route in the past

[12,17,18], and the influence or pressure of primary

social relationships (sexual partner, friends, family)

[5,14,19,20]. Among the reasons observed to explain the

adoption and maintenance of the smoked or sniffed

routes are the influence or pressure of the social environ-

ment (including fear of the social stigma of injection)

[6], concern about the negative health consequences of

injection [12], particularly fear of HIV infection [17,21]

and the market availability of high purity heroin for

smoking (in base form) [22] or sniffing (in salt form) at

competitive prices [20,21,23].Most of these factors have been detected in epidemio-

logical studies of association (individual or ecological),

without asking users directly about the reasons for initi-

ating, maintaining or giving up a particular route of

administration. Studies of association have made it possi-

ble to identify important factors that users may not per-

ceive, and to describe some of the social and demographic

characteristics of users adopting or changing to a given

route. Nevertheless, it is necessary to complete this view

with studies that directly explore the drug users opinions

and perceptions.

An in-depth exploration of drug transition routes inSpain is of particular interest because the AIDS epidemic

related with injecting drug use (mainly heroin) has been

one of the most severe of all the developed countries [24].

Furthermore, the spread of injected heroin use in Spain

occurred during a time of rapid transformation from an

authoritarian political regime to a democratic system,

with additional socio-economic problems, a situation

which may present many similarities to that of some east-

ern European countries today.

This study aims to identify the main reasons for select-

ing or changing to a specific route of heroin administra-

tion, and to explore their temporal trends. Its results

could be useful in guiding decisions on harm reduction

policy in countries that have undergone an experience

similar to that of Spain.

MATERIALS AND METHODS

Design and participants

A cross-sectional survey in three Spanish cities was car-

ried out between March and December 1995. The three

cities were selected based on the prevalence of the differ-

ent routes of administration of heroin [25]: in Barcelona

the injected route had predominated for some time; in

Madrid the smoked route had recently become the pri-

mary one; and in Seville the smoked route had predomi-

nated for a number of years. The initial sample consisted

of 909 users. All were both regular (life-time use of heroin15 times) and recent users (heroin used during past

30 days), and all had used heroin at least weekly at some

point in their lives. Three hundred and five participants

lived in Seville, 304 in Madrid and 300 in Barcelona. In

each city, approximately half the users were recruited in

drug treatment centres and the rest were recruited in the

street. All participants in the drug treatment sample had

begun treatment in 1995, and none of the heroin users

recruited outside the centres had been treated for heroin

addiction in the previous year. In selecting the treatment-

users all drug addiction treatment centres reporting to

the State Information System of Drug Abuse (SEIT inSpanish) during 1994 were included in the sample,

except for prison programmes. The total included 25 cen-

tres. In each city the sample in each centre was assigned

proportionally to the number of treatments reported to

the SEIT in the second quarter of 1994.

Users out of treatment were selected through targeted

sampling [26] and snowball sampling techniques [27].

Some 65.4% (299) were recruited directly by the inter-

viewers in areas where it was assumed there would be a

relatively high probability of finding them; 19.3% were

introduced or named by key informants and 15.3% were

named by other people interviewed (snowball sampling).The target areas for recruitment were meeting places for

users (68.9%), areas where drugs are sold (22.7%) and

others (8.4%). Services and settings which could have led

to a selection bias in terms of route of administration

(needle exchange programmes, pharmacies and others)

were not included. Key informants who named or intro-

duced users were friends or acquaintances of interview-

ers (42.0%), friends of the people interviewed (3.4%),

workers in treatment services (15.9%) and other people

who worked with drug users (38.6%). Snowball sampling

was used in an attempt to make the sample more repre-


3/12

Heroin administration routes during HIV epidemic 751


sentative by including people who might not be found in

traditional target areas (those more integrated into main-

stream society) as well as hidden networks of users. For

this purpose, each person interviewed who was selected

by targeted sampling or named by key informants was

asked to name up to four heroin users who met the inclu-

sion criteria and who, if possible, were not present in the

same target area where the person interviewed was

recruited. This technique was not as efficient as expected,

however, as most users did not name anyone (only 182

people were named); when they did, the people named

were often present in the same place, or it was difficult to

interview them because they could not be located, or they

did not keep their appointments. The 28 interviewers

were people who had privileged access to the target areas

and the centres where they worked because of their per-

sonal and professional contacts. Verbal informed consent

was obtained from all study participants. The sampling

methodology has already been described in detail else-where [8].

Information was obtained by personal interview. A

structured questionnaire was used, including pre-coded

questions for the following variables: socio-demographic

profile, current use of drugs, history of heroin use, evolu-

tion of usual route of heroin administration (URHA), his-

tory of injection of any drug, and HIV serological status.

The reasons for five behaviours related with the route of

administration were investigated separately) adopting

injection as the main (most frequent) route when usual

(weekly) heroin use started (first URHA); (b) changing

URHA to injection (to consider a change as URHA tran-sition, it had to be maintained for at least 30 days); (c)

never having injected drugs; (d) adopting the smoked or

sniffed route as first URHA; and (e) changing URHA to a

non-injecting route.

To investigate reasons for adopting the first URHA, a

closed list of factors or circumstances that might have

influenced the adoption of the initial URHA (injecting,

sniffing or smoking) was presented to each subject. For

each factor, the interviewee was asked to indicate

whether or not it had been present when the initial URHA

was adopted and, if so, what importance he/she gave to

that factor (very important, rather important, notvery important, not at all important). Two more lists of

suggested factors were presented, one for those whose

first URHA was injection and one for those whose first

URHA was either sniffing or smoking. Reasons for never

having injected any drug and for changing the URHA

were explored in a way similar to that described above. In

the latter case, if various changes had taken place, only

the most recent one was taken into account. Additional

questions were asked about whether the change in URHA

had been reached after a period of abstinence (and, if so,

how this period of abstinence had been attained), and

how the transition affected the amount of heroin used.

The interviewer offered a different set of factors depend-

ing on whether the change was to injection or to a non-

injecting route. The lists were developed based on a pre-

vious exploratory study. Nevertheless, to avoid excluding

any factors that may not have been included in the closed

list and to determine the importance given spontaneously

to certain reasons, subjects were asked in an open-ended

question to mention the three most important reasons for

adopting each of the five behaviours. This approximation

to the free-listing technique [28] as a method of prelimi-

nary exploration of the free and spontaneous discourse of

the person interviewed was carried out before presenting

the closed lists to avoid suggesting any particular reason

to the study subjects.

Data analysis

The analysis included 900 of the 909 subjects initiallyrecruited (305 in Seville, 298 in Barcelona and 297 in

Madrid). Nine people were eliminated because of major

inconsistencies in their replies about changes in the route

of administration.

The proportion of users who said that a specific factor

or circumstance had been present when adopting a first

URHA or changing URHA was calculated. We then cal-

culated the percentage of people who considered that fac-

tor rather or very important in adopting the behaviour

investigated. The highest non-response rate for any

behaviour investigated through a closed list was 3.0%.

The c2 test or c2 for trend was used to compare propor-tions. The null hypothesis was rejected when P< 0.05.

The reasons stated in the open-ended questions were

recorded and transcribed literally. Two investigators read

and classified these reasons and discrepancies indepen-

dently were resolved by consensus. The response rate for

each category identified were calculated. The proportion

of users who did not state spontaneously any reasons

varied between 2.0% and 4.4%.

The statistical analysis was performed with SPSS/

PC6.0 for Windows [29].

RESULTS

General characteristics of the sample

A detailed description of the sample has been published

elsewhere [8,30]. Most participants were male (81.8%),

aged 2535 (68.1%), single (72,6%), had fewer than

9 years of school education (77.7%) and did not have a

regular job (79.2%). Some 26.1% received most of their

income from illegal activities and another 23.3% from

marginal occupations. Almost half (46.2%) had been in


4/12



prison. The mean duration of heroin use was 10.2 years.

Ninety per cent used heroin on a daily or nearly daily

basis, 70.1% had used cocaine during the past

12 months and 48.5% had used it during the past

30 days.

With regard to the URHA, 51.7% were smokers,

41.7% were injectors and 6.6% were sniffers. Smokers

predominated in Seville (76.1%) and Madrid (70.7%),

and injectors predominated in Barcelona (77.3%), with

an increasing trend in the use of non-injecting routes in

all three cities. Forty-five per cent changed the URHA

during their drug use career, usually just once (32.3%).

With respect to the most recent change of URHA, 50.1%

changed to injection, 41.2% changed from injection to

smoking or sniffing and 8.7% changed between non-

injecting routes (from sniffing to smoking or vice versa).

Seventy-one per cent had injected drugs at least once,

with major variations among cities. Most subjects, when

injecting for the first time, used heroin (86.6%) and wereinjected by someone else (55.3% by a close friend, 8.2%

by a casual acquaintance, 3.3% by a family member and

2.4% by others). The proportion of those injected by

someone else was higher before 1987 (79%) than after-

wards. Women had been injected for the first time by their

sexual partner in a much larger proportion than men

(29.8% versus 1.3%, P< 0.00001), although the total

proportion of those injected by another person was not

significantly different from that of men (81.6% and 74%,

P= 0.13).

Reasons for choosing injection as first URHA

Of all the participants, 301 chose injection as the first

URHA. The most important reason was the influence of

the social environment; that is, the fact that either most

of their friends or their sexual partner also used this route

(Table 1). The importance of having an injector as a sex-

ual partner was much stronger among women: 50.0% of

women considered that this was a rather or very impor-

tant reason to adopt the injected route versus 10.2% of

men (P< 0.00001). These differences are due mainly to

the fact that when women started injecting they were

much more likely than men to have had a sexual partnerwho injected (52.2% versus 15.3%, P< 0.00001).

Among those whose sexual partner injected, 95.8% of

women and 66.7% of men (P= 0.02) considered that it

was an important reason for adopting the injected route.

The second reason was the belief that injected heroin has

a greater or better effect than smoking or sniffing (Table

1).

When temporal variations were analysed, a decreas-

ing but non-significant trend in the importance of the

influence of the social environment was observed in more

recent years. An increasing belief in injection as more

efficient than smoking or sniffing was also observed

(Table 1).

Reasons related to the market (unavailability of heroin

suitable for smoking or sniffing, or availability of low

purity heroin) were considered rather or very important

by 28.3% of subjects in the closed list (Table 1), but hardly

anyone mentioned this factor spontaneously in the pre-

vious open-ended question. In fact, only 1.4% mentioned

the availability of good heroin for injection as an impor-

tant reason.

Reasons for transitions from usually smoking or sniffing

heroin to injecting

Of all those interviewed, 202 changed from usually

smoking (104) or sniffing (98) to injecting in the most

recent URHA transition. Of these, 14.3% changed after a

period of abstinence lasting at least 1 month, achieved in

most cases without the assistance of treatment. After thechange of route, 62.4% of participants started using a

smaller quantity of heroin, 29.2% continued using the

same amount and 8.4% started using more.

The two main reasons for changing the URHA were

the superior effectiveness (better or greater effect) and

efficiency (same effect for less money) of injection com-

pared to other routes. Another relatively important rea-

son for the transition was the influence or pressure of the

social environment (Table 1). Having a partner who

injected was a more important reason for changing to the

injected route in women: 38.9% of women considered it

was a rather or very important reason versus only 7.2%of men (P< 0.00001). These differences are due mainly

to the fact that when women changed to injection they

much more frequently had a sexual partner who injected

than did men (50.0% versus 8.4%, P< 0.00001). Among

those who had an injecting sexual partner, 77.8% of the

women and 85.7% of the men considered that this fact

was important in explaining the change to injection, a

difference that was not statistically significant.

There were important differences in the reasons given

by those who changed from smoking and those who

changed from sniffing to injection. Smokers more said

frequently that the availability of better heroin for inject-ing was an important reason for the transition (30.6%

versus 13.5%, P= 0.005), and more often referred to the

belief that injecting was a more efficient route (62.2%

versus 45.6%, P= 0.03), whereas they were less influ-

enced by the social environment (31.6% versus 57.7%,

P= 0.0003).

In looking at temporal variations, it was found that

the influence of the social environment decreased signif-

icantly. It also appears that the importance of difficulties

in smoking or sniffing heroin due to intolerance or dis-

ease, the availability of better heroin for injecting than for


5/12



smoking or sniffing, and the belief in the superior effi-

ciency of injected heroin all increased significantly over

time (Table 1).

Market reasons (unavailability of good quality heroin

to smoke or sniff, decreased purity) were considered

important in explaining the change to injection for21.7% of users (Table 1), but these reasons were not

expressed in the previous open-ended question.

Reasons for never having injected drugs

Of the total sample, 262 subjects had never injected

drugs. The two major reasons for most people were con-

cern about negative health consequences (especially

fear of HIV and other infections, and fear of overdose)

and fear of blood or of the physical act of inserting a

needle in ones veins. More than half those interviewed

who had never injected chose reasons related to the

influence of the social environment or of having enough

money to buy heroin suitable for smoking or sniffing

(Table 2).

Few people expressed concern about a higher risk of

overdose from injection in the previous open-ended ques-tion. In fact, only 4.2% of those who had never injected

mentioned this reason.

Reasons for choosing smoking or sniffing as first URHA

Of the total sample, 599 users adopted a non-injecting

route (384 smoking and 215 sniffing) as the first URHA.

Most of those interviewed considered that concern about

negative health consequences (infections, overdose or

dependence) was the main reason for choosing the

smoked or sniffed route. The second most frequently cho-

Table 1 Importance given to reasons for adopting injection as first usual route of heroin administration (URHA) and for changing from

smoking or sniffing to injecting heroin.

% who considered this reason as rather or

very important in adopting injecting as first

URHAa

% who considered this reason

as ratheror very important in

changing from smokingor

sniffing to injectingb

Year of initiation of usual use of heroin/year of most recent

changeof main route of administration Total


6/12



sen reason was social influence or pressure, mainly by

friends or the sexual partner (Table 3). Similar to the case

of injection, the importance in absolute terms of having a

sexual partner who did not inject when selecting smoking

or sniffing as the initial route of administration was much

greater for women. Indeed, 47.0% of women considered

it was a rather or very important reason versus 8.6% of

men (P< 0.00001). These differences were due mainly to

the fact that when they adopted the smoked or sniffed

route, women were more likely than men to have a sexual

partner who used heroin by these routes (51.3% versus

10.9%, respectively, P< 0.00001). Among those who

Fear of health consequences of injection 95.8

Fear of infection by HIV or other agents 87.4

Belief that injecting implies higher risk of overdose 76.7

Belief that injecting implies much higher risk of dependence 58.0

Fear of blood or of inserting a needle in ones veins 89.3

Influence of social environment 59.5

Fear of losing sexual partner 37.4Fear of losing the majority of ones friends 29.7

Fear of being discovered and of losing ones job 24.8

Having enough money to buy heroin suitable for smoking or sniffing 54.1

Always having good heroin available for smoking or sniffing 20.2

Table 2 Reasons given as rather or very

important for never having injected

(n= 262).

Table 3 Importance given to reasons for adopting sniffing or smoking as first URHA, and for switching from injecting to smoking or sniffing

heroin.

% who considered thisreason as rather

or very important in adopting sniffing orsmoking as first URHAa

% who considered this reason

as ratheror very important in

switching frominjecting tosmoking/sniffingb

Year of initiation of usual use of heroin/year of most recent change

of main route of administration Total


7/12



had a sexual partner using the same non-injecting route,

91.7% of women and 78.8% of men (P= 0.10) consid-

ered that this was important in the adoption of the

smoked or sniffed route.

There were some differences between those whose

first URHA was the smoked route and those who sniffed.

Smokers more often feared HIV and other infections

(51.8% for smokers and 30.0% for sniffers, P< 0.0001),

believed that heroin used by the selected route was

more effective than injecting (23.5% versus 11.3%,

P= 0.0004) and believed that the risk of overdose was

lower (50.1% versus 36.5%, P= 0.002).

When analysing temporal trends, it was seen that the

importance of health reasons increased, in particular the

fear of HIV or other infections (Table 3). This phenome-

non was most notable among sniffers, among whom the

proportion of those who considered health reasons as

important increased from 43.8% before 1982 to 88.0%

in 199295 (significant linear trend, P= 0.0002), whilethe proportion of those who feared infection rose from

16.7% to 60.0% (P= 0.0003). The importance of having

a sexual partner who used the same route also increased.

The importance of market availability of good heroin suit-

able for smoking/sniffing decreased, in particular the

availability of high purity heroin (Table 3). Another sig-

nificant trend was seen among sniffers: the proportion of

those who said their belief that sniffed heroin has a better

effect than injected heroin was important in their choice

of first URHA dropped from 18.8% before 1982 to 0.0%

in 199295 (P= 0.02).

In the open-ended question the belief that smoking orsniffing implies a lower risk of overdose was rarely cited

(1.9%).

Reasons for transition from usually injecting heroin to

smoking or sniffing

Regarding their most recent change of URHA, 166 users

shifted from injecting to smoking (148) or sniffing (18).

Of these, 45.5% changed after a period of abstinence last-

ing at least 1 month, achieved in many cases without

help from any treatment centre. After the transition,

52.7% of participants began using larger quantities ofheroin, 21.0% continued using the same amount and

26.3% began using less.

The most important group of reasons for giving up

injecting were related with negative health conse-

quences. Fear of becoming infected or of finding out the

result of an HIV test was considered an important reason

for 61.4% (Table 3). The two next most frequently chosen

reasons were the difficulty of injecting because of poor

vein conditions and the influence of the social environ-

ment. The importance of difficulties in injecting was

higher among women than among men: 70.0% of

women considered this reason as important in changing

to a non-injecting route versus 37.0% of men

(P= 0.002). With regard to the influence of the social

environment, the importance of having a sexual partner

who smokes or sniffs heroin was greater among women:

30.0% of women considered that this was rather or very

important versus 13.2% of men (P= 0.05). These differ-

ences are due mainly to the fact that women more often

had a sexual partner who smoked or snif fed heroin when

they changed to that route (30.0% of women versus

16.2% of men, P= 0.13). Among those who had a sexual

partner who sniffed or smoked, 81.8% of men and 100%

of women considered that this fact was important when

changing to smoking or sniffing.

An increased capacity to buy heroin (because of

higher income or lower priced heroin) and the availability

of good quality and high purity heroin suitable for smok-

ing or sniffing were also important reasons (Table 3).

Being in prison at the time of the transition to smoking orsniffing was selected as an important reason by 10.2% of

those who adopted one of those routes.

The importance of having veins in poor condition as a

reason for changing to a non-injected route decreased

significantly over time. The importance of the belief that

smoking/sniffing implies a lower risk of overdose than

injecting also decreased, but this was not statistically sig-

nificant. Conversely, there was an increase over time in

the importance of social pressure, as well as the fear of

HIV infection or the knowledge of an HIV test result,

although these differences did not reach statistical signif-

icance (Table 3).The availability of heroin suitable for smoking or sniff-

ing was rarely mentioned spontaneously in the open-

ended questions (1.8%). Similarly, the belief that the risk

of overdose is lower from smoking or sniffing than

from injection was mentioned by only 4.8% of those

interviewed.

DISCUSSION

Factors influencing initiation of injecting or transition to

this route

The reasons for adopting injection initially as the URHA

were different from those chosen to explain the transition

to this route. The influence of the social environment was

most important in the choice of injection as the initial

route, whereas the transition to injection was more influ-

enced by the conviction that injecting heroin is more

effective or efficient than smoking or sniffing it.

Our data suggest that during the first years of the her-

oin epidemic in Spain (before 1987), the influence of

peers (friends, neighbours, school companions) or sexual

partners was the most important factor affecting the


8/12



adoption of injection as the initial route of heroin admin-

istration. In fact, 74.4% of those who chose injection as

the URHA between 1982-1987 considered that social

influence had been important in selecting this route; in

79% of cases they were injected for the first time by

another person, generally a close friend. From 1987

onwards, the spread of the smoked route resulted in a

decreasing proportion of heroin users who injected; thus

the weight of the social factor in the initiation of injecting

also decreased, although it remained important.

Our study and others [5,8,12,16] have found that

having an injector as a sexual partner can be decisive in

initiating or shifting to the injected route. Our study also

suggests that the effect of this factor is much stronger for

women than for men, mainly because the proportion of

women who have an injecting sexual partner is much

higher, and also because women give more importance to

this factor.

The belief that injection is more efficient or effectivethan other routes was the most important reason for the

transition to injecting. Other studies have found similar

results [13,14,19]. The importance that drug users

attribute to this reason reveals the unstable situation of

many smokers and sniffers with a high level of depen-

dence. In fact, this group may well act as a time-bomb

that could produce a flood of new injectors when the

appropriate conditions arise, such as restrictions in the

supply of heroin suitable for smoking, a rise in the price/

purity rate, etc. Although injecting is not necessarily an

inevitable consequence of heroin use [8,20], it has been

observed recently that the high frequency of smoking orsniffing is an important determinant of the transition to

injecting [12].

One factor not considered as important either in

adopting injection as the first URHA or in changing to

this route was the availability of free sterile syringes

through various programmes. These results are consis-

tent with evidence obtained by other methods

[5,16,31].

Several factors related to the choice of injection as the

initial route of heroin administration or the transition to

this route were not perceived spontaneously as relevant

by most users. One was the market factor, especially thecharacteristics of heroin supplied on the market (unavail-

ability of heroin suitable for smoking or sniffing, availabil-

ity of low purity heroin). The reason why the heroin

supply is not perceived spontaneously as an important

influence on the route of administration is not clear, but a

similar phenomenon probably occurs in the daily life of

people in the general population when they choose what

products to buy. Few people think of the influence of sup-

ply on their patterns of consumption and those who do

tend to overrate their freedom of choice. In any case,

although there is probably a strong association between

the supply and route of heroin administration, little evi-

dence exists about its direction or temporal sequence.

Were users forced to begin smoking because base heroin

replaced white heroin in the drug market, or did this

change occur because users demanded a product with

less risk? Probably both things happened, to some extent.

In any case, if characteristics of the heroin supply can

influence the route of administration, this implies that

changes in policies to control the drug supply may modify

the proportion of heroin users who inject. Similarly,

stronger control of drug trafficking could provoke an

increase in the price : purity ratio, and thus an increase in

the proportion of injectors. An ethnographic study car-

ried out in New York shows, however, that restrictions on

the heroin supply do not automatically or immediately

produce transitions to injection, and that users may have

multiple responses, such as starting a drug dependence

treatment, using other drugs, reducing or even eliminat-

ing use [23]. This phenomenon would in any case have tobe confirmed in other areas.

Factors influencing initiation of smoking/sniffing heroin

and transition to these routes

Contrary to what others have found [15,32], heroin

users in Spain state that the fear of health conse-

quences (infections, overdose, dependence) are the most

important reason why they have never injected heroin

or why they have changed to a non-injecting route.

Between 1982 and 1995 health reasons were most

important with regard to the adoption of smoking orsniffing as first URHA, although this did not occur dur-

ing the early years of the heroin epidemic. The most

important health reason, particularly in the last period

studied, was fear of HIV or other infections, or of hav-

ing an HIV test. Some authors suggest that the influ-

ence of the AIDS epidemic and of prevention

programmes might have induced safer drug use behav-

iour, including changes to non-injecting routes

[21,33,34]. This influence should have been especially

strong in Spain, where the prevalence of HIV infection

among injectors reached 40% to 75% [3,35] during the

1980s. Nevertheless, neither the intensity of the AIDSepidemic in the various Spanish regions nor the activi-

ties carried out to fight it can explain satisfactorily the

temporal and geographical trends in the different routes

of heroin administration. In fact, the transition to smok-

ing began before there was a public awareness that HIV

was transmitted by contaminated syringes, and the

regions with the highest prevalence of HIV among

injectors have always been (and still are) those with the

highest proportion of injectors among heroin users

[36]. Other health reasons for adopting a non-injecting

URHA were the belief that injecting heroin implies a


9/12



higher risk of overdose than smoking or sniffing and

that it also implies a greater risk of dependency. The

higher risk of dependency when injecting appears

rather doubtful in the light of some studies [12,37], par-

ticularly when compared with smoking. The deep-

rooted belief that injecting is a riskier route for depen-

dency may be explained by the results of several other

studies [38,39] and by the strong negative perception of

injecting compared to other routes of administration.

This belief may help to slow down the transition

towards injection, but may also make smoking and

sniffing appear of little consequence.

The second most important factor in switching from

injecting to smoking/sniffing was the difficulty of inject-

ing due to vein problems. Injecting in many different sites

of the body, as well as the emergence of vein problems,

seem to be more frequent among women than among

men [40], which is consistent with our studys finding

that women more often perceive difficulties in injectingbecause of vein problems as important in the transition to

a non-injected route.

The influence or pressure of the social environment

was also an important reason for stopping injection.

Norms and group attitudes against injectors may exist

nowadays, because they are perceived as losers in very

poor health [5]. This factor may have become increas-

ingly important in recent years, which is logical consid-

ering that the proportion of heroin users who adopt non-

injecting routes has been continuously on the rise.

An aversion to needles or to injections was a very

important reason for never having injected drugs. Thisfactor has been pointed out by other authors [7], and it

probably varies by social or ethnic group. Some injectors

have even mentioned that they felt this aversion before

they started injecting drugs and that they were only able

to overcome it when another person injected them while

they were looking the other way [41]. Finally, contrary to

what might be expected, being in prison seems to have

been more important in the case of the transition to non-

injecting routes than the other way around.

Some factors were not perceived spontaneously as

important by most users with regard to the choice of

smoking or sniffing as the first URHA. Most strikingamong these was market conditions (availability of her-

oin suitable for smoking or sniffing, availability of high

purity heroin, low price of heroin). The influence of this

factor in the case of smoking seems obvious, since it is

practically impossible to smoke heroin if one cannot

obtain base heroin. In New York, the decrease in the

price : purity ratio of heroin seems to have coincided with

an increase in its use by sniffing [42]. Moreover, the avail-

ability of high purity heroin suitable for smoking has

probably favoured the fact that a large proportion of her-

oin sniffers in the United States have not shifted to injec-

tion [17,43]. Similarly, the belief that the smoked or

sniffed route involves a lower risk of overdose was rarely

cited spontaneously by any of those interviewed. The

open-ended questions did not reveal any important rea-

sons that were not already included in the categories of

the closed lists for each of the five types of behaviour

studied.

Implications for intervention and research

Acting on the factors that influence injection is a high pri-

ority to avoid or reduce the principal health problems

associated with heroin use. Given the high risk of depen-

dence related with non-injecting use, it is also important

to limit the spread of heroin sniffing and smoking.

This study suggests that multiple reasons often exist

for adopting a given route of heroin administration,

which result from the interaction of individual, social and

market factors. Although some factors are unlikely to bemodified by social or health-related actions, innovative

interventions should be designed to prevent drug users

from injecting [10]. For instance, recognition of the influ-

ence of peers and sexual partners means that initiating or

shifting to injection may be partly avoided by acting on

and through the peer group and its leaders. One immedi-

ate idea is to act directly on sniffers and smokers with a

high risk of transition by making them more conscious of

the risks of injection, helping them to develop skills to

confront offers of injection, and helping them to avoid or

reduce contacts with injector networks. Peers and sexual

partners may have a decisive influence on the adoption ofinjection, and are often the people from whom injectors

receive their first heroin injection. Thus, it is also neces-

sary to focus on those who are already injecting, who are

more visible and accessible than non-injectors, to prevent

them from injecting or promoting injection among their

non-injecting sexual partners and friends [10]. There is

evidence that these interventions are feasible, acceptable

and effective [9]. In any case, given the strong influence of

sexual partners on the choice of the route of administra-

tion, a high priority would be to act on sexual partners

with an injecting and a non-injecting member.

The perception of health risks related to injectingseems to have acted as a strong determinant in choosing

the smoking or sniffing route, and in the transition to

non-injecting. In this regard, it is also necessary to make

heroin users more conscious of the risk of hepatitis C

infection and of overdose.

The fact that many users may change to injection

because of its greater efficiency and effectiveness means

that smokers and sniffers with a high level of dependence

are exposed to an important risk of injecting, albeit spo-

radically. Consequently, they should be a target group for

prevention and treatment programmes (with adequate


10/12



doses of oral opioids), and should be particularly aware

that the risk of HCV and HBV infection at the beginning of

an injection career are very high [11,17,44], as is the risk

of overdose for sporadic injectors [4547].

Other factors, such as the characteristics and price of

heroin available on the market, are difficult to control,

and little effort has been made to design and implement

such interventions. We believe it is possible to choose

market control options that may minimize the risk of

injecting. It has been proposed recently that control

efforts should be concentrated on countries and black

market sectors that produce and distribute salt heroin

[10].

Study limitations

This study explores the subjective reasons chosen from a

series of lists provided by study investigators of reasons forinitiating or changing routes of heroin administration.

These reasons may not coincide completely with the

objective determinants of such behaviours. In addition it

is evident that, even though we made a previous explor-

atory study to ensure inclusion of the reasons given in the

usual discourse of drug users, the formulation of closed

replies is always a reductionist approximation of reality.

The inclusion in the interview of an open-ended question,

posed neutrally to avoid biasing the response in any way,

palliates this limitation to some degree and has enriched

the results. Both methods offer similar and complemen-

tary information, and no strong inconsistencies wereseen. As in many studies of illegal drug use, questions

could be raised about how representative our sample was

based on two main factors: the non-probabilistic nature

of the subsample of street-users and ignorance of the pro-

portion of treatment-users in the whole population of

heroin users. To minimize the problems derived from the

first factor we worked with a large sample, seeking to

include as diverse a selection of users as possible. The sec-

ond factor is unlikely to introduce major bias because the

differences between the general characteristics of treat-

ment-users and street-users were small. The results of

this study may also be limited by recall bias, or may reflectthe dominant perception regarding the phenomenon

under consideration more than the subjects true reasons

at the time. Moreover, most of the questions about the

reasons for different types of behaviour focused on the

usual route of heroin administration, which did not allow

us to explore, for instance, the reasons for occasional

injection perceived by those who usually smoked or

sniffed. Finally, this study focused on heroin users (which

in Spain include the great majority of injectors); there-

fore, we do not know to what extent these results may

apply to other drugs such as cocaine.

ACKNOWLEDGEMENTS

The fieldwork for this study was financed through FIS

project 94/1527, and the data analysis through FIPSE

project 3035/99. We would like to thank Aurelio Daz

and Mila Barruti for their co-operation in the question-

naire design; Valentn Mrquez and Juan Gamella(Madrid) for their collaboration in the definition of set-

tings and the selection of interviewers. We also wish to

thank the City and Regional Plans on Drugs that facili-

tated access to drug treatment centres (Lluis Torralba,

Josep Mara Suelves, Carlos Mateo, Emiliano Martn

and Fernando Arenas), as well as all drug treatment

centre co-ordinators. The views expressed in this work

are solely the responsibility of the authors and are not

necessarily shared by the institutions in which they

work.

REFERENCES

1. Sporer, K. A. (1999) Acute heroin overdose. Annals of Inter-

nal Medicine, 130, 584590.

2. Gossop, M., Griffiths, P., Powis, B., Williamson, S. & Strang,

J. (1996) Frequency of non-fatal heroin overdose: survey of

heroin users recruited in non-clinical settings. British Med-

ical Journal, 313, 402.

3. Bravo, M. J. & De la Fuente, L. (1991) Epidemiologa de la

infeccin por VIH en los usuarios de drogas por va

parenteral [Epidemiology of HIV infection among IVDUs].

Publicacin Oficial de Sociedad Espaola Interdisciplinaria de

SIDA, 2, 335342.4. Stimson, G. V. (1996) Drug injecting: the public health

response in the next decade. Addiction, 91, 10981099.

5. Van Ameijden, E. J. & Coutinho, R. A. (2001) Large decline

in injecting drug use in Amsterdam, 19861998: explana-

tory mechanisms and determinants of injecting transitions.

Journal of Epidemiology and Community Health, 55, 356363.

6. Swift, W., Maher, L. & Sunjic, S. (1999) Transitions

between routes of heroin administration: a study of Cauca-

sian and Indochinese heroin user s in south-western Sidney,

Australia. Addiction, 94, 7182.

7. Strang, J., Griffiths, P. & Gossop, M. (1996) Heroin smoking

by chasing the dragon: origins and history. Addiction, 91,

673683.

8. De la Fuente, L., Barrio, G., Royuela, L., Bravo, M. J. & TheSpanish Group for the Study of the Route of Heroin Admin-

istration (1997) The transition from injecting to heroin

smoking in three Spanish cities. Addiction, 92, 17331744.

9. Hunt, N., Stillwell, G., Taylor, C. & Griffiths, P. (1998) Eval-

uation of a brief intervention to prevent initiation into

injecting. Drugs: Education, Prevention and Policy, 5, 185

194.

10. Hunt, N., Griffiths, P., Southwell, M., Stillwell, G. & Strang,

J. (1999) Preventing and curtailing injecting drug use: a

review of opportunities for developing and delivering route

transition interventions. Drug and Alcohol Review, 18, 441

451.

11. Van Ameijden, E. J. C. & Coutinho, R. A. (1998) Maximum


11/12



impact of HIV prevention measures targeted at injecting

drug users. AIDS, 12, 625633.

12. Van Ameijden, E. J. C., Van den Hoek, J. A. R., Hartgers, C. &

Coutinho, R. A. (1994) Risk factors for the transition from

noninjection drug use and accompanying AIDS risk behav-

ior in a cohort of drug users. American Journal of Epidemiol-

ogy, 139, 11531163.

13. Crofts, N., Louie, R., Rosenthal, D. & Jolley, D. (1996) The

first hit: circumstances surrounding initiation to injecting.

Addiction, 91, 11871196.

14. Gamella, J. (1994) The spread of intravenous drug use and

AIDS in a neighbourhood in Spain. Medical Anthropology

Quarterly, 8, 131160.

15. Casriel, C., Des Jarlais, D. C., Rodrguez, R., Friedman, S. R.,

Stepherson, B. & Khuri, E. (1990) Working with heroin

sniffers: clinical issues in preventing drug injection.Journal

of Substance Abuse Treatment, 7, 110.

16. Des Jarlais, D. C., Casriel, C., Friedman, S. R. & Rosenblum,

A. (1992) AIDS and the transition to illicit drug injection

results of a randomised trial prevention programme. British

Journal of Addiction, 87, 493498.

17. Mathias, R. (1999) Heroin snorters risk transition to injec-

tion drug use and infectious disease. NIDA Notes, 14, 111.18. Neaigus, A., Miller, M., Friedman, S. R., Hagen, D. L., Sifa-

neck, S. J., Ildefonso, G. & Des Jarlais, D. C. (2001) Potential

risk factors for the transition to injecting among non-

injecting heroin users: a comparison of former injectors and

never injectors. Addiction, 96, 847860.

19. Casriel, C., Rockwell, R. & Stepherson, B. (1988) Heroin

sniffers: between two worlds. Journal of Psychoactive Drugs,

20, 437440.

20. Sotheran, J. L., Goldsmith, D. S., Blasco, M. & Friedman, S.

R. (1999) Heroin sniffing as self-regulation among inject-

ing and non-injecting heroin users. Journal of Drug Issues,

29, 401422.

21. Hamid, A., Curtis, R., McCoy, K., McGuire, J., Conde, A.,

Bushell, W., Lindenmayer, R., Brimberg, K., Maia, S.,Abdur-Rashid, S. & Settembrino, J. (1997) The heroin epi-

demic in New York City: current status and prognoses.Jour-

nal of Psychoactive Drugs, 29, 375391.

22. De la Fuente, L., Saavedra, P., Barrio, G. Royuela, L. & Vice-

nte, J. (1996) Temporal and geographic variations in the

characteristics of heroin seized in Spain and their relation

with the route of administration. Drug and Alcohol Depen-

dence, 40, 185194.

23. Andrade, X., Sifaneck, S. J. & Neaigus, A. (1999) Dope sniff-

ers in New York City: an ethnography of heroin markets and

patterns of use.Journal of Drug Issues, 29, 271298.

24. European Centre for the Epidemiological monitoring of

AIDS (EuroHIV) (2001) HIV/AIDS Surveillance in Europe.

End-year report 2000, no. 64. Saint Maurice, France:EuroHIV.

25. Delegacin del Gobierno para el Plan Nacional sobre Drogas

(DGPNSD) (1994) Sistema Estatal de Informacin sobre Toxi-

comanas (SEIT), Informe 1993 [State Information System on

Drug Abuse, Report 1993]. Madrid: Ministerio de Justicia e

Interior.

26. Watters, J. K. & Biernacki, P. (1989) Targeted sampling:

options for the study of hidden populations. Social Problems,

36, 416430.

27. Hartnoll, R., Griffiths, P., Taylor, C., Hendrick, V., Blanken,

P. & Nolimal, D. (1997) Handbook on Snowball Sampling.

Strasbourg: Pompidou Group, Council of Europe.

28. Ort, A. (1990) La apertura y el enfoque cualitativo o

estructural: la entrevista abierta semidirectiva y la dis-

cusin en grupo [The qualitative or structural focus: the

semi-directed interview and group discussion]. In: Alianza

Editorial, ed. El anlisis de la realidad social. Metodos y tcnicas

de investigacin [The Analysis of Social Reality: Research Meth-

ods and Techniques], pp. 171203. Madrid.

29. Norussis M. J. (1993) SPSS for Windows: Advanced Statistics,

Release 6.0. Chicago: SPSS Inc.

30. Barrio, G., De la Fuente, L., Royuela, L., Daz, A., Rodrguez-

Artalejo, F. & The Spanish Group for the Study of the Route

of Drug Administration (1998) Cocaine use among heroin

users in Spain: the diffusion of crack and cocaine smoking.

Journal of Epidemiology and Community Health, 52, 172

180.

31. Vlahow, D. & Junge, B. (1998) The role of needle exchange

programs in HIV prevention. Public Health Reports, Supple-

ment 1, 7580.

32. Sibthorpe, B. & Lear, B. (1994) Circumstances surrounding

needle use transitions among injection drug users: implica-

tions for HIV intervention. International Journal of the Addic-

tions, 29, 12451257.

33. French, J. F. & Safford, J. (1989) AIDS and intranasal her-

oin. Lancet, i, 1082.34. Des Jarlais, D. C., Friedman, S. R. & Ward, T.P. (1993) Harm

reduction: a public health response to the AIDS epidemic

among injecting drug users. Annual Review of Public Health,

14, 413450.

35. Hernndez-Aguado, I., Avino, M. J., Prez-Hoyos, S.,

Gonzlez-Aracil, J., Riz-Prez, I., Torrella, A., Garca de la

Hera, M., Belda, M., Fernndez, E., Santos, C., Trullen, J. &

Fenosa, A. (1999) Human immunodeficiency virus (HIV)

infection in parenteral drug users: evolution of the epidemic

over 10 years. International Journal of Epidemiology, 28,

335340.

36. De la Fuente, L., Lardelli, P., Barrio, G., Vicente, J. & Luna, J.

D. (1997) Declining prevalence of injecting as main route of

administration among heroin users treated in Spain, 19911993. European Journal of Public Health, 7, 421426.

37. Barrio, G., De la Fuente, L., Lew, C., Royuela, L., Bravo, M. J.

& Torrens, M. (2001) Differences in severity of heroin

dependence by route of administration: the importance of

length of heroin use. Drug and Alcohol Dependence, 63, 169

177.

38. Gossop, M., Griffiths, P., Powis, B. & Strang, J. (1992) Sever-

ity of dependence and route of administration of heroin,

cocaine and amphetamines. British Journal of Addiction, 87,

15271536.

39. Smolka, M. M. & Schmidt, L. G. (1999) The influence of her-

oin dose and route of administration on the severity of the

opiate withdrawal syndrome. Addiction, 94, 11911198.

40. Darke, S., Ross, J. & Kaye, S. (2001) Physical sites amonginjecting drug users in Sydney, Australia. Drug and Alcohol

Dependence, 62, 7782.

41. McBride, A., Pates, R. M., Arnold, C. & Ball, N. (2001) Nee-

dle fixation, the drug users perspective: a qualitative study.

Addiction, 96, 10491058.

42. Strang, J., Des Jarlais, D. C., Griffiths, P. & Gossop, M. (1992)

The study of transitions in the route of drug use: the route

from one route to another. British Journal of Addiction, 87,

473483.

43. National Institute on Drug Abuse (NIDA) (2000) Epidemio-

logical Trends in Drug Abuse, vol. I. Proceedings of the Commu-

nity Epidemiological Work Group, June 2000, pp. 3250.

Bethesda, MD: NIDA.


12/12

2003 Society for the Study of Addiction to Alcohol and Other Drugs Addiction 98 749 760


44. Garfein, R. S., Vlahov, D., Galai, N., Doherthy, M. C. & Nel-

son, K. E. (1996) Viral infections in short-term injection

users: the prevalence of the hepatitis C, hepatitis B, human

immunodeficiency, and human T-lymphotropic viruses.

American Journal of Public Health, 86, 655661.

45. Brugal, M. T., Barrio, G., De la Fuente, L., Regidor, E.,

Royuela, L. & Suelves, J. M. (2002) Factors associated with

non-fatal heroin overdose: assessing the effect of frequency

and route of heroin administration.Addiction, 97, 319327.

46. Van Haastrecht, H. J. A., Van Ameijden, E. J. C., Van Den

Hoek, J. A. R., Mientjes, G.H.C., Bax, J. S. & Coutinho, R. A.

(1996) Predictors of mortality in an Amsterdam cohort of

human immunodeficiency virus (HIV) positive and HIV

negative drug users. American Journal of Epidemiology, 143,

380391.

47. Tagliaro, F., De Battisti, Z., Smith, F.P., & Marigo, M. (1998)

Death from heroin overdose: findings from hair analysis.

Lancet, 351, 19231925.

Reasons for Selecting an Initial Route of Heroin

Documents

Transcript of Reasons for Selecting an Initial Route of Heroin