Addiction Research and Theory

October 2006, 14(5): 461–473

A drug monitoring system: Keeping a finger

on the pulse by triangulation of qualitative
and quantitative methods

H. VAN DE MHEEN, M. COUMANS, C. BARENDREGT,

& A. VAN DER POEL
Addiction Research Institute Rotterdam, 3021 DM Rotterdam, The Netherlands

(Received 8 August 2005; accepted 5 June 2006)

Abstract
The aim of monitoring is to keep a finger on the pulse. In this article, we focus on a drug
monitoring system (DMS): a local research system, continuously collecting data on hard
drugs, drug users, and related issues. The methodology of the drug monitoring system
combines qualitative and quantitative methods, i.e., community fieldwork, interviews
with key informants and a bi-annual drug user survey. The system has been developed
inductively: theory and methods were developed from daily practice. The present article
aims to provide a scientific and methodological foundation for a DMS. We incorporated
questions with respect to internal and external validity. The following strategies have been
applied to meet the criteria of ‘trustworthiness’: triangulation, prolonged engagement,
persistent observation, member checks, peer debriefing, negative case analyses, thick
description, and reflexive journal. Methodological triangulation used in a complementary
model appears to be an important tool.
Keywords: Monitoring, (hard) drugs scene, triangulation

Introduction
Monitoring can be seen as a method of signalizing trends and developments.
An important characteristic of monitoring is its power to combine an
epidemiologic and social science approach. It demands a targeted combination
of quantitative methods, including the repeated measurement of standardized
core indicators, as well as qualitative methods on questions dictated by the
empirical social reality at the current time and place. Monitoring gathers

Correspondence: H. van de Mheen, Addiction Research Institute Rotterdam, Heemraadssingel

194, 3021 DM Rotterdam, The Netherlands. Tel: þ31 10 4253366. Fax: þ31 10 2763988.
E-mail: vandemheen@ivo.nl

ISSN 1606-6359 print/ISSN 1476-7392 online ß 2006 Informa UK Ltd.

DOI: 10.1080/16066350600903310
462 H. van de Mheen et al.

continuous information on a specific phenomenon, over an extended period of

time. The aim is to keep a finger on the pulse of, in our case, trends and
developments in the use of (hard) drugs.
By monitoring phenomena over a longer period of time, trends can be spotted.
This yields important information for policy, prevention programs, and
treatment. However, the benefit of a monitoring system is not always obvious.
Although a monitoring system can be running for a long period of time without
any obvious effect in society, such a system can still be fruitful.
To have its use for policy, a monitoring system must be a ‘management
information system’, based on scientific methods. This implies that information
is gathered, ordered, and analyzed on a systematic and continuous basis. In
addition, frequency tables and simple analyses must be continuously available,
information must be updated frequently, and qualitative data (e.g., ethnographic
fieldwork) must be coded and ordered continuously. Theoretically, there is a
cyclic process: a monitor provides information for a more evidence-based local
drug policy (also with respect to treatment and prevention), and the local drug
policy influences the research issues and data collection.

The drug monitoring system

This article focuses on a specific kind of monitor, i.e., a drug monitoring system.
Griffiths, Vingoe, Hunt, Mounteney and Hartnoll (2000) described the use of
drug monitoring systems, based on work for the European Monitoring Centre for
Drug and Drug Addiction (EMCDDA). They gave a worldwide overview of
existing drug monitoring systems and described their use in terms of operational
level, structure, function, range of data sources, and investigation methods.
One of the systems described by Griffiths et al. is the Dutch Drug Monitoring
System (DMS) (Van de Mheen, 2000).
The DMS is a local research system (carried out in several localities/regions in
the Netherlands) continuously collecting both quantitative and qualitative data
about drugs, drug users, and related issues. The target population is the group of
hard drug users. Its small-scale design allows the use of ethnographic research
methods to provide insight into situations, developments, and trends. The results
inform local policy makers, drug-treatment agencies, and others about the
dynamics of the local drug scene. The community-based research methods of the
monitor allow a quick reaction to developments and give possibilities to effectively
describe the effects of policy implementation. The continuous character and
the integration of different research methods strengthen the study above the level
of just an ad hoc measurement. The main research topics are: characteristics of the
drug users’ group, buying and using of drugs (type of drugs, location of drug use,
route of administration, mechanisms of self-regulation), work and income,
housing, and physical and mental health of drug users (Addiction Severity Index
problems) (Hendriks, Meer, & Blanken, 1991). Important is the focus of
the DMS on the drug user in his or her natural environment, or ‘the daily life’.
Within this natural environment the meaning of a described phenomenon can be
 Drug monitoring: Triangulation of methods 463

interpreted in its context (setting) (Zinberg, 1984). Data for the DMS are mainly
collected by means of reports from community fieldworkers, interviews with key
informants, and a bi-annual drug user survey. These methods will be described
later.
The DMS is incorporated in a report of Netherlands Organization for Health
Research and Development about monitoring systems on alcohol and drugs in
The Netherlands (Cruts, Lempens, Ketelaars, Van Laar, & Van de Mheen,
2004). Compared to other monitors in The Netherlands it is unique in its
combination of qualitative and quantitative methods. In addition, most monitors
are targeted at the general population. The DMS is one of the few monitors that is
targeted at hard drug users.
In theory the DMS only evaluates local drug policy. Therefore, in the
Netherlands Drug Monitor (Van Laar, Cruts, Verdurmen, & Van Ooyen, 2006),
it plays a role in signalizing trends in certain local situations that may also occur at
other places or at the national level. The DMS evaluates national drug policy as
far as the national drug policy influences the local situation.

Aim of the article

The DMS method was designed in the 1990s (Hendriks, Blanken, Adriaans,
& Vollemans, 1994), and further developed during daily practice. Experience
of researchers and fieldworkers guided the choice of the most appropriate
methods, combining sources of information, and creating new methods of
sampling and data collection. These give the DMS a substantial practical
power, which has proven useful for local policy (Van de Mheen, 1999).
However, the scientific basis of the monitoring system, especially questions with
respect to internal and external validity, has never been documented. The
present article aims to provide a scientific and methodological foundation for
a DMS. Research questions are: (1) what are the main methodological
elements in the system? and (2) what is the scientific basis of the system in
terms of validity?

Methodology of the DMS

The DMS combines three methods, which are complementary: a survey among
drug users (quantitative method), a network of key informants (qualitative
method), and community fieldwork (qualitative method). These methodological
elements are shown in Figure 1 and described later.

Survey among drug users

Initially, information is gathered by surveys among drug users themselves.
Depending on the study population and specific research questions the survey
uses different methods, e.g., written questionnaires and (semi) structured
interviews. The surveys are carried out every 2 years among 100–200
respondents, depending on the size of the local population. Sampling methods
464 H. van de Mheen et al.

Community Key
fieldwork informants

Check results

- Topics - Topics
- Hypotheses - Hypotheses
- Sampling

Survey

Figure 1. Complementary methods used in the DMS.

used are snowball sampling, network sampling, and targeted sampling (for more
details, see Coumans & Spreen 2003; Barendregt, Van der Poel, & Van de
Mheen, 2005).
The questionnaire is partly based on the Addiction Severity Index (Hendriks
et al., 1991). The duration of one interview ranges from 30 min to 1.5 h. Analyses
are conducted by means of SPSS software and comprise simple statistical
procedures, such as ‘descriptives’, ‘frequencies’, ‘cross tabulations’ and ‘t-tests’.
For every bi-annual survey a new sample is taken.
An informed consent procedure was incorporated. We handed out written
information about the research project to the respondents. Informed consent was
obtained by filling in the questionnaire. A signature was required for the response
reward. Since we did not ask the name and address of the respondent, privacy and
anonymity was guaranteed.

Key informants
A network of key informants reports regularly about phenomena and develop-
ments in and involving drug use, on the basis of their work, position, and
membership of a group, or expert experience. They belong to representatives of
care providers, local policy makers, police, different groups of users, and groups
related to users (e.g., drug dealers). Each year one or two rounds of formal
interviews take place (individual or in focus groups), in which the general
monitoring items are complemented with a certain focus, e.g., ‘mobility’ or
‘health’. In addition, key informants report on an informal basis about their
findings based on observations of and conversations with members of the study
populations.
The network consists of about 10–15 key informants. The interviews last on
average 1 h and are done with the help of a pre-structured checklist. All interviews
are tape-recorded and typed out verbatim. The most informative interviews,
in the sense of focus or perspective of the key informants, are coded and analyzed
with the software program Ethnograph (Seidel, 1998).
 Drug monitoring: Triangulation of methods 465

Community fieldwork
Community fieldworkers are drug users or ex-drug users who report about events
and phenomena of the drug scene. They are familiar with the common concepts
and operations from their knowledge and experience with the scene and have easy
access to the targeted research group. They carry out a great part of the DMS
fieldwork, and interview drug users for the bi-annual survey. Community
fieldworkers play an important role in the DMS. They: (1) describe and localize
the population of hard drug users on the local level, (2) make and keep contact
with (groups of) users, or to speak generally, develop a ‘research alliance’ between
hard drug users and the research team, and (3) are helpful in interpretation of the
research data (Blanken & Barendregt, 1998; Blanken, Barendregt, & Zuidmulder,
2000; Van de Mheen, 2000).
Drug users’ participation means that some of them are selected on the basis of
their background, knowledge, and experience with drug use and the concurrent
lifestyle. Community fieldworkers are hired on a freelance basis and, after a short
training, the development of their ethnographic skills is subject to continuous
attention of the researcher. The community fieldworkers and the researcher
observe and interact with the drug users in their own natural environment.
After each field session these observations and interactions are written down in
the so-called field notes. In order to distinguish personal thoughts and questions
from pure observations or interactions, the field note also contains reflections on
these experiences. After writing up the entire field note, the text is structured
according to the three elements of observation, interaction, and reflection.
Each field note is evaluated with the researcher and at the end of each field note
the fieldworker is required to formulate at least one question about the content
of the note. These questions serve as guidance for future fieldwork sessions.
The field notes are entered in Microsoft Word and analyzed with the software
program Ethnograph (Seidel, 1998). For the coding of the data a standardized
code list has been developed that can be adapted for locally determined purposes.

Combination of methods used in DMS

The DMS is a research project in which qualitative methods are essential.
Drug users can be viewed as a ‘hidden population’. The term ‘hidden
population’ euphemistically refers to ‘those who are disadvantaged and
disenfranchised: the homeless and transient, chronically mental ill, high school
dropouts, criminal offenders, prostitutes, juvenile delinquents, gang members,
runaways, and other ‘‘street people’’ – those we are all aware of to one degree or
another, yet know so little about’ (Lambert, 1990, p. 1). The study by Pearson
(2001) among inner city drug users in London (UK) is a good example of
research among such a hidden population.
Ethnography is one type of qualitative research and is by nature fundamental
and exploratory, and useful for topics about which little is known. It gives the
researcher the opportunity to understand the world as seen by the respondent
within the context of the respondent’s everyday life (Strunin, 2001).
466 H. van de Mheen et al.

According to Sterk (2003, p. 3): ‘the applied nature of drug research is ideal for
critical ethnography, in which the individuals who are part of the study, through
their participation, gain a better understanding of their reality and also insights for
potential change’.
Qualitative methodology has made an enormous development in the last
decades. This development started with the publication of Glaser and Strauss
(1967), which explains the ‘grounded theory’, followed by more publications
about qualitative methods and analysis (e.g., Chenitz & Swanson, 1986;
Denzin, 1978; Miles & Huberman, 1984). Qualitative methods are useful in the
exploratory stages of the research project, where they will help to clarify or even set
the research question, provide conceptualization, and generate hypotheses for
later research. They may also be used to interpret, qualify or illuminate the
findings of quantitative research and to test hypotheses. Qualitative methods are
very useful to help policy makers understand the context in which policy will be
implemented (Murphy, Dingwall, Greatbatch, Parker, & Watson, 1998).
Qualitative research methods are applied (at least in the DMS) in different
ways, combined with quantitative methods. First, qualitative research is used
as a precursor to quantitative research. Second, it can be used to explain
unanticipated or inconclusive findings from quantitative studies. Third,
qualitative research is used to generate hypotheses (Murphy & Dingwall,
1998). Qualitative methods can also be used independently, i.e., not combined
with quantitative methods; however, that is not the case in the DMS.
The importance of qualitative methods in understanding the dynamic nature of
the drug scene and relating problems has become evident from the literature
(Lambert, Ashery, & Needle, 1995). Also in public health and epidemiology
the monopoly of quantitative research has been replaced by a growing attention
for qualitative research (Bourgois, 2002; Jones, 1995; Mackenbach, 2001;
Savage, 2000). Classical epidemiology can be seen as a ‘black box’: it
produces figures that do not provide insights in the mechanisms of causality
(Mackenbach, 2001). In (social) drug research quantitative and qualitative
researchers have a great deal to offer each other. Bourgois (2002) formulates this
as follows: ‘The absence of a dialogue between epidemiological and qualitative
researchers – especially ethnographers who engage in cultural anthropology’s
version of participant–observation methods – is a failure from the perspective
of both pragmatic positivism of public health and also the critical theory of
anthropology’.
Recently, Neale, Allen and Coombes (2005) emphasized the importance of
qualitative research methods within the addictions. They showed that the
strengths of qualitative addiction research are many and diverse, and that they
have particular advantages in studying hidden and hard-to-reach populations.

Validity or ‘trustworthiness’
According to Lincoln and Guba (1985), we use different terms to refer to the
naturalistic (or qualitative) equivalents of the validity criteria that are used
 Drug monitoring: Triangulation of methods 467

in quantitative research. All these different terms refer to the overall criterion
of ‘trustworthiness’.
The equivalent of internal validity in quantitative research can be found in
credibility: the probability that the findings will be found to be credible and
to be approved by the ‘constructors of the multiple realities being studied’
(Lincoln & Guba, 1985, p. 301). The equivalent of external validity can be
described as transferability, or: to what extent can we transfer the findings to
other contexts? Strictly speaking it is impossible to reach external validity by
qualitative research (Lincoln & Guba, 1985). However, we argue that
transferability is achieved by the generalization of theoretical statements,
which may guide policy makers but have to be tested through application
in other contexts. Dependability in qualitative research accounts for the
quantitatively used criterion of reliability in terms of (in) stability and
‘factors of phenomenal or design induced changes’ (Lincoln & Guba, 1985
p. 300). Confirmability is the qualitative equivalent of objectivity; it
emphasizes the characteristics of the data, instead of the characteristics of the
researcher.
In the DMS, according to Lincoln and Guba (1985), several strategies are
performed to meet the trustworthiness criteria. To meet the criterion of credibility
and dependability, we used triangulation, prolonged engagement, persistent
observation, member checks, peer debriefing, and negative case analyses.
To account for the criterion of transferability, we applied ‘thick description’,
and confirmability was accounted for by writing a reflexive journal.
These techniques are described later; we start with triangulation since it applies
to the DMS as a whole.

Triangulation
Triangulation is the combination of different sources leading to better knowledge
or, according to Nutbeam (1999), it gathers evidence from different sources: the
more consistent the evidence from different sources, the more the validity of the
results can be presumed. In other words: triangulation is a method applying more
than one approach to answer the same question.
Commonly, the following forms of triangulation are distinguished (a.o. Denzin,
1978):
– Triangulation of data: different sources of information are used deriving from,
e.g., client dossiers, minutes of meetings, publications, and interviews with key
informants.
– Triangulation of researchers: more than one researcher is involved in data
collection and analysis. This may be useful if researchers have different
theoretical or methodological backgrounds.
– Triangulation of methods: different methods (usually quantitative and
qualitative) are used to study the research subject, e.g., focus groups,
individual interviews, participant observations.
468 H. van de Mheen et al.

Triangulation does not represent a single methodological concept but has

different understandings, of which the two most commonly used are that of
triangulation as mutual validation, and triangulation as the integration of different
perspectives on the same investigated phenomenon (complementary model)
(Kelle, 2001).
Denzin (1978) and Maxell (1998) define triangulation in the sense of checking
the validity and reliability of the findings (mutual validation). As opposed to
relying on one single form of evidence or perspective as a basis for findings,
multiple forms of diverse and redundant types of evidence are used to reduce
the risk of systematic distortions inherent in the use of only one method
(Maxwell, 1998). Methodological triangulation refers to a synergetic effect:
evidence from different sources increases the confidence in research findings.
Or, in other words: a hypothesis that has survived a series of tests with different
methods could be regarded as more valid than a hypothesis tested only with one
single method (Denzin, 1978).
However, other authors warn against the use of triangulation in this way
(Morse, 1991; Murphy et al., 1998). Research results would only be usable for
mutual validation if they relate to the same phenomenon, since only then can
different results correctly be interpreted as indicators of validity problems. If one
method is related to a different aspect of the same phenomenon, or even a
different phenomenon, then one should expect a different result. In that case the
methods are complementary (Kelle, 2001). Morse (1991, 2000, 2003)
emphasizes the complementary model. This is the use of at least two methods,
usually qualitative and quantitative, to address the same research topic.
Or, in other words: information from one data source can provide background
information for another data source, or control or supplement data from another
source. When a single method is inadequate, triangulation is used to ensure
that the most comprehensive approach is taken to solve a research problem.
According to Morse, methodological triangulation can be classified as simulta-
neous or sequential. Simultaneous means that qualitative and quantitative
methods are used at the same time: there is limited interaction between datasets
during the data collection; findings complement each other at the end of
the study. Sequential means that results of one method are essential for planning
the next method. Qualitative methods are completed before the quantitative
methods are implemented or vice versa. Methodological triangulation is a
method of obtaining complementary findings that strengthens research results
and contributes to theory development rather than duplicating findings
(Morse, 1991).
In our DMS the model of complementarity is more appropriate than the model
of mutual validation. We combine qualitative and quantitative methods to
supplement each other. Neither qualitative nor quantitative methods used as
a single method provides sufficient information to describe or understand the
complete social process. The different methods relate to different aspects of the
same phenomenon. The phenomenon under study is the scene of hard drug
users. The information gathered from key informants (qualitative interviews),
 Drug monitoring: Triangulation of methods 469

community fieldwork (qualitative participant observations), and a survey among

drug users (quantitative questionnaires) elucidates different views (e.g., the
perspective of the drug users, the perspective of environment) on the same
research subject.
In the DMS we had no a priori theoretical framework. The process is primarily
inductive, and qualitative methods have the highest priority. An inductive process
means that the theory is developed qualitatively and complemented by
quantitative methods. The process is more sequential than simultaneous; this is
shown in Figure 1. For example, information from key informants and
community fieldwork influences the focus of the bi-annual survey among drug
users. This fits with the original ideas of Glaser and Strauss (1967) that results
obtained by qualitative methods are particularly well suited to develop theory.
The results can guide survey researchers in both sample construction and
research question development (Watters & Biernacki, 1989).

Other techniques in the DMS

Prolonged engagement implies considerable investment (above all, time) to know
the culture of the drug scene and to develop trust. To prevent distortions,
in the DMS the fieldworkers interact with drug users in their own environment
(e.g., low-threshold services or dealing addresses). Fieldworkers had to decide
whether they preferred to be a ‘fly on the wall’, or a ‘wasp’ constantly asking
(sometimes annoying) questions. A potential pitfall of this latter technique was
‘going native’; to over-identify with the research subject and thus lose the research
perspective. In the DMS we tried to prevent this, for example by not taking part
in an interest group for drug users. Persistent observation implies that only things
that are relevant are incorporated in the study. In the DMS it was important to
develop antennas to distinguish between relevant and irrelevant issues. Therefore,
we used the basic questions of the DMS (see Introduction), and the specific
questions the fieldworker was required to formulate at the end of each field note.
A possible pitfall is ‘premature closure’, or focusing too soon on specific issues.
In the DMS, this meant maintaining an open mind for new phenomena, such as
an increasing number of youngsters using crack. Another example is that of the
transition to mobile phone (Barendregt, van der Poel, & Van de Mheen, 2006),
that was firstly reported by community field workers. Also the feeling among drug
users that municipal by-laws with respect to gathering and public drinking
(accompanied by an increase of penalties) victimize drug users, was a perspective
brought by drug users. In addition an example of ‘premature closure’ is that
care-givers focus on problems and not on in depth details of the use itself.
The researcher should be aware of the problem oriented view of these
key-informants. According to Lincoln and Guba (1985, p. 314) ‘the member
check is the most crucial for establishing the credibility of the findings’. Data,
analytic categories, interpretations, and conclusions are tested with members
of the research population. Reconstructions made by the researcher must be
recognizable to the respondents. In the DMS, this meant that drug users were
470 H. van de Mheen et al.

offered an opportunity to respond to the research findings. This was

accomplished by, e.g., organizing a focus group for key informants and drug
users in which the findings were presented. In terms of validity ‘lay’ knowledge of
users is always geographically restricted. It is the task of the researcher to put this
in a wider perspective. Peer debriefing means that in all phases of the research the
researcher discusses his or her thoughts, ideas, analyses and conclusions with a
‘peer’: a (colleague) researcher who tries to play the devil’s advocate. The goal of
this technique is to explicate important aspects (e.g., working hypotheses) that
might otherwise remain implicit in the head of the researcher. In the DMS,
researchers from different cities played the role of each other’s peer and an
external advisor was employed to (temporarily) play the role of a peer without
judging the researcher.
Negative case analysis means that, similar to statistical tests, findings are more
and more generalized to all cases during the process, by actively seeking deviant
cases. After studying these cases, the hypothesis can be (re)adjusted until it
covers all cases without exception. In the DMS this technique implied that we
tried to describe all categories of drug users, including all possible variations.
Thick description (Geetz, 1973) holds that descriptive data describing certain
themes in a pithy way are incorporated in the report; the intention is that
interpretations and findings be strengthened by these data thus made more
credible. Such consideration about the data enables an outstander to judge
whether or not transfer to another context is appropriate. In the DMS, we used
quotations from interviews and field notes to support descriptions of the topic of
interest.
Finally, the reflexive journal is a general strategy, which can be applied to
account for all criteria and can also be used in quantitative research. There are
several possible applications of this technique: e.g., the researcher makes
notes about his or her own behaviour and personal thoughts and considerations,
and/or keeps a record with methodological decisions. In practice this meant that
during the DMS survey a logbook was kept, and the field notes allowed room for
personal reflections on the observations.

Conclusions
This article presents a theoretical framework and justification for the method-
ology of (particularly) drug monitoring. We defined a monitoring system as a
‘management information system’, based on scientific methods. The aim is to
keep a finger on the pulse, and thereby signal trends and developments.
The methodology of the DMS combines qualitative and quantitative methods,
i.e., community fieldwork, interviews with key informants and a bi-annual drug
user survey. Qualitative research methods, especially ethnographic research,
appear to be appropriate methods for research among ‘hidden populations’, and
thus for the DMS.
We showed that, although our DMS was not developed from an a priori theory,
there is a sound scientific methodological basis for the results it has yielded and
 Drug monitoring: Triangulation of methods 471

will yield. The system has been developed inductively: theory and methods were
developed from daily practice. Questions with respect to internal and external
validity are incorporated in the system, by applying the criteria of Lincoln and
Guba (1985): credibility, transferability, dependability, and confirmability,
all referring to the overall criterion of ‘trustworthiness’. We showed that in a
DMS it is possible to include strategies to meet the criteria of ‘trustworthiness’:
triangulation, prolonged engagement, persistent observation, member checks,
peer debriefing, negative case analyses, thick description, and a reflexive journal.
Especially the technique of triangulation is an important tool in a DMS.
In our system we used methodological triangulation in a complementary
model. That means that information gathered with one method is used to
supplement the results from other methods, and to provide background
information for, or to study in depth the results from, other methods. We believe
that this model is more appropriate in drug monitoring than the validation model,
in which the aim is to duplicate findings (checking validity) by use of other
methods.
Generalizability or transferability is shown to be of importance in qualitative
research where the theoretical statements are applicable to other target
populations. In our case, theoretical concepts of marginalization and hardening
which are important in the drug scene under study (Coumans, Knibbe,
& Van de Mheen, in press), can also be applied to populations of, e.g., alcoholics
or homeless people.
We conclude that it is possible to develop an integrated drug information
system, in which data collected from different sources and by different methods
are subsequently evaluated, that has a scientific methodological basis. The results
inform local policy makers about the dynamics of the local drug scene. The drug
monitoring system is not targeted to evaluate the effectiveness of local drug
policy. Although it can give global indications whether a local drug policy
reached its goals, it can never give evidence for a causal relationship. A DMS can
provide policy makers and care providers with accurate and relevant
information about developments and trends in (hard) drug use, which serves
a more evidence-based local drug policy, also with respect to treatment and
prevention.

References
Barendregt, C., van der Poel, A., & van de Mheen, H. (2005). Tracing selection effects in three
non-probability samples. European Addiction Research, 11, 124–131.
Barendregt, B., van der Poel, A., & van de Mheen, H. (2006). The rise of the mobile phone in the
hard drug scene in Rotterdam. Journal of Psychoactive Drugs, 38, 77–87.
Blanken, P., & Barendregt, C. (1998). Veranderingen in de aanbod – en vraagzijde van de
illegale drugsmarkt in Rotterdam: Enkele resultaten van het Rotterdams Drug Monitoring
Systeem 1995–1997. (Changes in supply and demand on the illegal drug market. Some
results from the Rotterdam Drug Monitoring System 1995–1997). IVO bulletin, 1(1), 1–11,
[in Dutch].
472 H. van de Mheen et al.

Blanken, P., Barendregt, C., & Zuidmulder, L. (2000). Community fieldwork: Bringing users into
research action. In: Understanding and responding to drug use: The role of qualitative research.
Lisbon: EMCDDA.
Bourgois, P. (2002). Anthropology and epidemiology on drugs: The challenges
of cross-methodological and theoretical dialogue. International Journal of Drug Policy, 13,
259–269.
Chenitz, W. C., & Swanson, J. M. (1986). From practice to grounded theory. Qualitative research
in nursing. California: Addison-Wesley Publishing Company.
Coumans, A. M., & Spreen, M. (2003). Drug use and the role of homelessness in the process
of marginalization. Substance Use & Misuse, 38, 311–338.
Coumans, A. M., Knibbe, R. K., & van de Mheen, H. (in press). Street level effects of local drug
policy on marginalisation and hardening: An ethnographic study among chronic drug users.
Journal of Psychoactive Drugs.
Cruts, A. A. N., Lempens, A. L. F., Ketelaars, A. P. M., van Laar, M. W., & van de Mheen, H.
(2004). Monitoring van alcohol en drugs onder de loep. Een onderzoek naar opbrengst
en lacunes van informatiesystemen op het gebied van alcohol en drugs. (Monitoring of
alcohol and drugs on the map: A study on outcome and lacunas of information
systems in the field of alcohol and drugs). Utrecht/Rotterdam, Trimbos Institute/IVO, 2004
[in Dutch].
Denzin, N. K. (1978). The research act. Chicago: Aldine.
Geetz, C. (1973). Thick description: Toward an interpretative theory of culture. In C. Geertz (Ed.),
The interpretation of cultures. (pp. 3–30). Selected Essays. New York: Basic Books.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative
research. New York: Aldine.
Griffiths, P., Vingoe, L., Hunt, N., Mounteney, J., & Hartnoll, R. (2000). Drug information
systems, early warning, and new drug trends: Can drug monitoring systems
become more sensitive to emerging trends in drug consumption? Substance Use & Misuse, 35,
811–844.
Hendriks, V. M., Meer, C. W., & Blanken, P. (1991). De Addiction Severity Index: ASI (R).
Handleiding bij training en afname (The Addiction Severity Index (R). Manual for training and
administration). Rotterdam: IVO [in Dutch].
Hendriks, V. M., Blanken, P., Adriaans, N. F. P., & Vollemans, L. C. (1994). Continue
gegevensverzameling over druggebruik: de opzet van een locaal drug monitoring system (DMS).
(Continuous data collection about drug use: The design of a local drug monitoring system
(DMS)). Rotterdam: IVO [in Dutch].
Jones, R. (1995). Why do qualitative research? British Medical Journal, 311, 2.
Kelle, U. (2001). Sociological explanations between micro and macro and the integration of
qualitative and quantitative methods. Forum: Qualitative Social Research [on line journal] 2(1).
http://qualitative-research.net/fgs/fgs-eng.htm.
Lambert, E. Y. (Ed) (1990). The collection and interpretation of data from hidden populations.
NIDA Research Monograph 98. Rockville: NIDA.
Lambert, E. Y., Ashery, R. S., & Needle, R. H. (Eds) (1995). Qualitative methods in drug abuse
and HIV research. NIDA Research Monograph 157. Rockville: NIDA.
Lincoln, Y.S., & Guba, E. G. (1985). Naturalistic inquiry. Newbury Park, CA: Sage Publications.
Mackenbach, J. P. (2001) Paradigma, succes en toekomst van de maatschappelijke gezondheidszorg
(Paradigm, success and future of public health). TSG, 79, 522–529 [in Dutch].
Maxell, J. A. (1998). Designing a qualitative study. In Leonard Blickman & Debra J. Rog (Eds),
Handbook of applied social research methods. Thousand Oaks: Sage.
Miles, M. B., & Huberman, A. M. (1984). Qualitative data analysis. A source back of new methods.
Newbury Park, CA: Sage Publications.
Morse, J. M. (1991). Approaches to qualitative-quantitative methodological triangulation.
Nursing Research, 40, 120–123.
Morse, J. (2000). Qualitative generalisibility. Qualitative Health Research, 9, 5–6.
 Drug monitoring: Triangulation of methods 473

Morse, J. M. (2003). Principles of mixed methods and multimethod research design. In

A. Tashakkori & C. Teddlie (Eds), Handbook of mixed method. Thousand Oaks: Sage.
Murphy, E., Dingwall, R., Greatbatch, D., Parker, S., & Watson, P. (1998). Qualitative research
methods in health technology assessment: A review of the literature. Health Technology
Assessment, 2, 1–273.
Murphy, E., & Dingwall, R. (1998). Qualitative methods in health services research. In N. Black,
J. Brazier, F. Fitzpatrick & B. Reeves (Eds), Health services research methods. London: BMJ
Publications.
Neale, J., Allen, D., & Coombes, L. (2005). Qualitative research methods within the addictions.
Addiction, 100, 1584–1593.
Nutbeam, D. (1999). Evaluatie van oorzak en gevolg in gezondheidsbevorderingsprojecten:
Aanpassing van onderzoeksmethoden aan interventiemethoden (Evaluating cause and
effect in health promotion programs: Matching research methods to intervention programs)
TSG, 77, 15–23 [in Dutch].
Pearson, G. (2001). Normal drug use: Ethnographic fieldwork among an adult network of
recreational drug users in inner London. Substance Use & Misuse, 36, 167–200.
Savage, J. (2000). Ethnography and health care. British Medical Journal, 321, 1400–1402.
Seidel, J. (1998). Appendix E. Qualitative data analysis. In: The Ethnograph V5.0. A program for
the analysis of text based data. Canada: Qualis research associates.
Sterk, C. E. (2003). Drug research: Ethnographies or qualitative works. International Journal of
Drug Policy, 14, 115–118.
Strunin, L. (2001). Assessing alcohol consumption: Developments from qualitative research
methods. Social Science Medicine, 53, 215–226.
Van Laar, M. W., Cruts, A. N., Verdurmen, J. E. E., Van Ooyen, M. M. J. (Eds) (2005). National
Drug Monitor. Annual Report 2004. Utrecht: Bureau NDM.
Van de Mheen, D. (1999). In: U.S.- Netherlands Addiction Workshop and Binational Symposium
on drug Abuse, Addiction Research and Innovation, October 19–20 1999. Den Haag:
Netherlands Research and Development Program on Substance Use and Addiction/US
National Institute on Drug Abuse (NIDA).
Van de Mheen, D. (Ed.) (2000). De Rotterdamse Drugsscene onder de loep. Resultaten van 5 jaar
Drugs Monitoring System in Rotterdam (Focus on the Rotterdam Drugs scene. Results
of a 5-year Drug Monitoring System in Rotterdam). IVO, Reeks Monitor Alcohol en Drugs,
nr 5. Rotterdam: IVO [in Dutch].
Watters, J. H., & Biernacki, P. (1989). Targeted sampling: Options for the study of hidden
populations. Social Problems, 36, 416–430.
Zinberg, N. E. (1984). Drugs, set and setting; the basis for controlled intoxication. New Haven:
Yale University Press.