Substance Abuse

ISSN: 0889-7077 (Print) 1547-0164 (Online) Journal homepage: https://www.tandfonline.com/loi/wsub20

Improving medical students’ attitude toward

patients with substance use problems through
addiction medicine education

Astri Parawita Ayu, Margot van der Ven, Eva Suryani, Natalia Puspadewi,
Satya Joewana, Elisabeth Rukmini, Cor de Jong & Arnt Schellekens

To cite this article: Astri Parawita Ayu, Margot van der Ven, Eva Suryani, Natalia Puspadewi,
Satya Joewana, Elisabeth Rukmini, Cor de Jong & Arnt Schellekens (2020): Improving medical
students’ attitude toward patients with substance use problems through addiction medicine
education, Substance Abuse, DOI: 10.1080/08897077.2020.1732512

To link to this article: https://doi.org/10.1080/08897077.2020.1732512

View supplementary material

Published online: 27 Feb 2020.

Submit your article to this journal

Article views: 20

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=wsub20
SUBSTANCE ABUSE
https://doi.org/10.1080/08897077.2020.1732512

ORIGINAL RESEARCH

Improving medical students’ attitude toward patients with substance use

problems through addiction medicine education
Astri Parawita Ayu, MD, PhDa,b , Margot van der Ven, MScb, Eva Suryani, MDa, Natalia Puspadewi, MD,
MMedEda, Satya Joewana, MDa, Elisabeth Rukmini, PhDa, Cor de Jong, MD, PhDb , and
Arnt Schellekens, MD, PhDb,c
a
School of Medicine and Health Sciences, Atma Jaya Catholic University of Indonesia, Jakarta, Indonesia; bNijmegen Institute for Scientist-
Practitioners in Addiction, Nijmegen, the Netherlands; cDepartment of Psychiatry, Radboud University Medical Centre, Nijmegen, the
Netherlands

ABSTRACT KEYWORDS
Background: Patients with addiction often encounter negative attitudes from health care professio- Attitude; perception;
nals, including medical doctors. Addiction medicine training might improve medical students’ atti- addiction medicine
tudes toward patients with addiction problems and change the way they think about addiction. education; medical students
We evaluated the effect of comprehensive addiction medicine training on students’ attitudes and
illness perceptions and explored which perceptions are most relevant for attitude development.
Methods: In a quasi-experimental non-randomized study, fourth-year students (n ¼ 296) partici-
pated in either addiction medicine training (intervention) or one of three other blocks (control).
We used the Medical Condition Regards Scale to measure attitudes and the Illness Perception
Questionnaire Addiction version for perceptions. We analyzed the effect of the intervention using
repeated measures MANOVA. The contribution of illness perception to attitude was explored in
the intervention group using linear regression analysis. Results: Addiction medicine training
improved students’ attitudes toward patients with addiction, compared to the control group. After
the training, students expressed a less demoralized perception, a stronger perception of a coher-
ent understanding of addiction, addiction as a cyclical condition, and attributed addiction more to
psychological factors, compared to the control group. In the intervention group, attitude and
emotional representation before training and illness coherence after the training were associated
with attitude after the training. Conclusions: Addiction medicine training is effective in improving
medical students’ attitudes toward patients with addiction and changing their illness perceptions
of addiction. The development of an understanding of addiction might be particularly relevant for
attitude improvement. These findings underscore the relevance of addiction medicine training as
part of medical curricula and argue for including aspects related to attitude development in
the curriculum.

Introduction cardiovascular problems with cocaine consumption, and pul-

monary disorders with marijuana smoking.3,6–8 All psychi-
Addiction is a chronic relapsing condition, involving bio-
atric disorders occur more commonly among patients with
logical, psychological and social aspects.1 It is manifested in
substance use disorders, particularly anxiety disorders,
compulsive behaviors, such as using a substance or gam-
depressive disorders and attention deficit hyperactivity dis-
bling. The worldwide prevalence of illicit substance addic-
order (ADHD).9 Thus, addiction is a significant public
tion is estimated at around 0.6% (0.4–0.9%).2 About 1% of
the global all-cause Disability-Adjusted Life Years (DALYs) health problem, and all physicians will encounter patients
are due to illicit substance addiction.3 The most common with addiction-related health problems in clinical practice.
substance use disorders however, are alcohol and tobacco Despite the importance to appropriately address addic-
use disorder. Alcohol and tobacco use account for 132.6 mil- tion-related problems, physicians often express negative atti-
lion and 148.6 million DALYs worldwide, respectively.4,5 tudes toward patients with addiction.10–12 Many of them
Addiction is often accompanied by physical and psychi- evaluate treating patients with addiction as unpleasant, unre-
atric complications. For example, hepatitis C and HIV infec- warding, and beyond their remit.10,13 Studies among health
tions are associated with injecting substance use, professionals (physicians, psychiatrists, psychologists, nurses,

CONTACT Astri Parawita Ayu, MD, PhD [email protected] School of Medicine and Health Sciences, Atma Jaya Catholic University of Indonesia,
Jl. Pluit Raya No. 2 Jakarta 14440, Indonesia.
Supplemental data for this article can be accessed on the publisher’s website.
ß 2020 Taylor & Francis Group, LLC
2 A. P. AYU ET AL.

social workers) and medical students have shown a more perceptions of addiction in medical students’ attitude devel-
negative attitude toward patients with substance use opment after addiction medicine training.
problems and addiction than that toward patients with, for
instance, diabetes, pneumonia, dementia, panic disorders, Methods
and depression.10,14 Several authors suggest that this nega-
Study design
tive attitude affects the quality of care for these patients.12,15
Indeed, patients with addiction report that they often experi- This is a quasi-experimental non-randomized study with pre
ence discrimination, includes receiving inadequate treatment and post measurements. Addiction medicine training was
when accessing healthcare.16,17 the intervention and three other elective pieces of training
The attitude of health professionals is among others asso- (health care entrepreneurship, palliative care, medical educa-
ciated with their perception of a certain condition.18 For tion) were the control condition.
example, medical doctors are more reluctant in treating
patients with addiction, when they believe that patients can
Participants
control their substance use themselves. Importantly, physi-
cians express widely different perceptions of addiction. A total of 296 fourth-year medical students of Atma Jaya
Some doctors perceive addiction as a disease, while others Catholic University of Indonesia participated in this study.
believe that addiction is a choice or way of coping with Some participants were recruited in 2014 (n ¼ 108), from
life,13 a moral weakness,19 or a social problem.20 Such per- three elective blocks: addiction medicine (n ¼ 40), health-
ceptions are thought to influence the quality of care for care entrepreneurship (n ¼ 35) and palliative care (n ¼ 33).
patients with an addiction.15,21 Others were recruited in 2016 (n ¼ 188), from four elective
Attitude and perception are, therefore, two distinct con- blocks: addiction medicine (n ¼ 46), healthcare entrepre-
cepts that have some overlapping aspects. The former neurship (n ¼ 47), palliative care (n ¼ 48), and medical edu-
focuses on the feelings of medical doctors and their behavior cation (n ¼ 47). Students were not randomly allocated to
toward working with patients with a particular medical con- the block since they could enroll in the block of their pref-
dition, whereas the latter depicts their ideas and cognitions erence. However, some students might not have been
about the illness in question. Both concepts are considered placed in their block of preference due to limited space
to influence doctors’ clinical behaviors toward the patients availability.
and can be altered by education.
A large body of studies has shown that addiction medi-
cine training in undergraduate medical education can Intervention
improve knowledge, skills, and attitudes concerning addic-
In Indonesia, medical education is divided into two levels:
tion among medical students.22–24 Such studies involved
preclinical and clinical. The Atma Jaya Catholic University
various types of addiction medicine training, ranging from
of Indonesia provided the above mentioned elective blocks
several hours of a workshop25 to a 3-week structured mod-
for fourth-year medical students at the pre-clinical level.
ule on addiction medicine.26 These studies generally show
Each elective block runs for 5 weeks. Students at the pre-
improvement of medical students’ attitudes after addiction
clinical level, including in these electives, do not engage in
medicine training,22,23 including alcohol,27 nicotine,24 and
substance use in pregnancy.23,24 any clinical work.
A major drawback of these studies is that they commonly The Addiction medicine block covered addiction topics
lack rigorous designs, for example, including a control ranging from theoretical aspects (e.g., epidemiology, neuro-
group, pre-post measurements, and well-validated assess- biology, etc.) to more clinical aspects, such as assessment
ment tools.22,23 Furthermore, no studies explored the effects and treatment, and comorbidities (see Supplemental
of addiction medicine training on illness perceptions of Material 1). The topics were provided as lectures, group dis-
addiction in relation to attitude development. Only one cussions, problem-based learning, and case-based learning.
study assessed the effect of a structured, comprehensive Students also learned to interview patients with addiction in
addiction medicine curriculum of several weeks on under- recovery, as a clinical skills training. The block offered field
graduate medical students’ attitudes toward addiction.26 trips to national addiction treatment centers. Students spent
However, this study lacked a control group and did not approximately 40 h per week to do activities in the block,
apply well-validated assessment tools. including self-study.
Here, we investigate the effect of comprehensive 5-week
elective addiction medicine training on undergraduate med- Instruments
ical students’ attitudes toward patients with addiction and
their illness perceptions of addiction, compared to a control The Medical Condition Regard Scale
group. Specifically, we tested the hypotheses that students in The Medical Condition Regard Scale (MCRS) intends to
the addiction medicine block (1) developed more positive measure attitude toward addicted patients. This is a one-
attitudes toward patients with addiction, and (2) changed dimensional scale, which measures the degree of regard
their perceptions of addiction, compared to those in the toward patients with the medical condition under study.
other blocks. Furthermore, we explored the role of The MCRS was developed among medical students using
 SUBSTANCE ABUSE 3

diverse medical conditions, including heartburn, pneumo- emotional representations, a high score reflects a perception
coccal pneumonia, depression, and intravenous drug use.14 of the emotional consequence of addiction.
It consists of 11 items on a six-point Likert scale (1: strongly The attribution scale reflects health professionals’ ideas
disagree to 6: strongly agree). Therefore, the total score about causes of addiction, which consists of four subscales:
ranges from 11 to 66. Higher scores reflect more positive psychological attribution, risk factors, smoking/alcohol, and
attitudes toward a specific condition, but there is no cutoff overwork. The psychological attribution section consists of
value for high versus low scores.14 items such as stress or worry and thinking negatively about
The MCRS has good psychometric properties (coefficient life. The risk factors attribution section includes items such
a ¼ 0.87, test–retest reliability ¼ .84).14 The MCRS was also as a germ or virus, diet or eating habits, and accident or
able to detect improvement of students’ attitudes toward injury as potential causes for addiction. High scores on a
depression, before and after a psychiatric clinical rotation subscale reflect a stronger belief that a factor causes addic-
(mean score: 48.3 and 52.1, respectively, p < .001). In add- tion. Since illness perception is a valence neutral concept,
ition, the MCRS has been used successfully to evaluate atti- representing personal beliefs about an illness36,38 the score
tude development among medical27–29 and other health of the IPQ-A, being either high or low, does not indicate a
professional students30–34 in previous studies. good or bad perception.

The Illness Perception Questionnaire Addiction version Procedure

Perceptions of addiction were measured with the Illness
Perception Questionnaire Addiction version (IPQ-A). This Questionnaires were distributed in the classroom on the first
questionnaire is an adaptation of the revised version of day of each elective block for the baseline measurement and
the Illness Perception Questionnaire (IPQ-R), which is on the last day for follow-up. All participants were asked to
mostly used to evaluate patients’ perceptions about their fill in questionnaires anonymously. Students were informed
illness.35 The IPQ-R was developed based on the illness about the study, and they provided written informed con-
representative, a concept to understand an illness based sent beforehand. The ethics committee of the School of
on cognitive (identity, timeline, consequences, control/cure, Medicine of the Atma Jaya Catholic University of Indonesia
and cause) and emotional components.35,36 The IPQ-A granted ethical clearance for this study (Format J, 7
focuses on what health professionals think about the con- October 2013).
dition addiction and consists of two domains: perception
(37 items) and attribution (15 items). A five-point Likert Analysis
scale (strongly disagree to strongly agree) is used for
each item. We analyzed all data using IBM SPSS software, version
The psychometric study of the IPQ-A revealed that it has 24.0.39 First, we used multivariate analysis of variance
eight subscales for the perception domain and four subscales (MANOVA) to explore differences in baseline attitudes and
for the attribution domain, which reflect similar factor struc- perceptions of addiction between the three blocks. The base-
tures with those of the IPQ-R.37 Moreover, the reliability of line scores were dependent variables and the four blocks
most of its scales was good (emotional representation: 0.88 (addiction medicine, healthcare entrepreneurship, palliative
and illness coherence: 0.86) and acceptable (consequences: care, and medical education) were the independ-
0.77, smoking-alcohol: 0.78, demoralization: 0.81, and psy- ent variables.
chological attribution: 0.82). Only five of them had poor Second, we analyzed the effect of the intervention (addic-
(treatment control: 0.53 and timeline cyclical: 0.63) and tion medicine block) on attitudes and perceptions of addic-
questionable (timeline chronic: 0.67, patient control: 0.67, tion using repeated measures MANOVA. This analysis uses
and risk factors: 0.71) reliability.37 In addition, the IPQ-A a within-subject approach, meaning that individual changes
was able to differentiate illness perception of students from were also tracked, by comparing pre-post measures within
different countries (Indonesia and the Netherlands) and subjects. The dependent variables were the attitude and per-
from various majors (medicine, psychology, and educa- ception of subscale scores. The within-subjects variable was
tional science).37 the time (two levels: baseline and follow-up) and the
The perception subscales are: demoralization, a high score between-subjects variable was the block (addiction medicine,
reflects a demoralized perception of addiction; timeline versus healthcare entrepreneurship, palliative care, and med-
chronic, a high score reflects a perception of a chronic time- ical education). We used Helmert contrast analysis, in order
line of addiction; consequences, a high score reflects a per- to analyze the difference between the intervention and the
ception of severe consequences of addiction on daily life and control group.
social function; personal control, a high score reflects a per- Finally, we performed explorative linear regression analy-
ception that the patient can control addiction; treatment ses to evaluate the predictors of the attitude after training
control, a high score reflects a perception that a treatment (follow-up) in the intervention group. The attitude follow-
will be helpful in managing addiction; illness coherence, a up was the dependent variable (outcome). The correlations
high score reflects a perception of having a coherent under- between the follow-up attitude and all perception subscales
standing of addiction; timeline cyclical, a high score reflects (baseline and follow-up) were explored. Perception subscales
a perception of a cyclical timeline of addiction; and that significantly correlated with the attitude at follow-up
4 A. P. AYU ET AL.

Table 1. Characteristics of participants.

Addiction medicine Healthcare entrepreneurship Palliative care Medical education
(n ¼ 86) (n ¼ 82) (n ¼ 81) (n ¼ 47 )
Female, n (%) 58 (67.4) 52 (63.4) 57 (70.4) 29 (61.7)
Age, mean ± SD 20.56 ± 0.63 20.54 ± 0.65 20.68 ± 0.74 20.64 ± 0.61

were included as predictors in the linear regression analysis Additionally, we did separate post hoc analyses concern-
using the stepwise method. ing the attitude changes within each intervention and con-
trol group. We found a significant effect of time in both
groups [F(1, 85) ¼ 11.232, p ¼ .001 and F(1, 209) ¼ 18.67,
Results p < .001), respectively]. These results suggest an improved
In 2014, in total 140 questionnaires were handed in the pre- attitude of students in the intervention group and a decline
measurement and 153 in the post-measurement (70% and in the control group.
75% response rate, respectively). Among these, 108 complete
questionnaires from both pre and post measurements were The effect of addiction medicine block on illness
analyzed. In 2016, 188 questionnaires were handed in (100% perceptions of addiction
response rate), both in pre and post-measurement. All ques-
tionnaires were filled in completely, and thereby they were A main effect of time on perceptions of addiction was
included in the analysis. There were no significant differen- observed, indicating that perceptions changed over time.
ces in gender and age between participants of the interven- These perceptions were Emotional Representation (F(1,
tion and control group. Table 1 describes the demographic 292) ¼ 5.06, p ¼ .03), Demoralization (F(1, 292) ¼ 6.11,
characteristics of the participants. p ¼ .01), Illness Coherence (F(1, 292) ¼ 14.49, p < .001),
Timeline Chronic (F(1, 292) ¼ 5.97, p ¼ .02), and Timeline
Cyclical (F(1, 292) ¼ 8.97, p < .01). There was also a main
Baseline effect of block on Demoralization (F(3, 292) ¼ 7.19,
At baseline, there were no differences in attitude, while only p < .001), Illness Coherence (F(3, 292) ¼ 6.84, p < .001),
the perception of Illness Coherence was significantly differ- Consequences (F(3, 292) ¼ 2.78, p ¼ .04), and Timeline
ent between addiction medicine and all other blocks Chronic (F(3, 292) ¼ 3.84, p ¼ .01).
There was an interaction between time and block for sev-
together (p ¼ .04). The Illness Coherence score of addiction
eral subscales (see Table 2). Further analysis with Helmert
medicine block was slightly lower than that of the healthcare
Contrast showed that the change of Demoralization
entrepreneurship, the palliative block, and the medical edu-
(p < .001), Illness Coherence (p < .001), Timeline Cyclical
cation block. This finding indicated that students who
(p ¼ .01), Psychological (p ¼ .02) subscales in Addiction
enrolled in addiction medicine block perceived themselves
Medicine block were significantly different to the
as having a less coherent understanding of addiction, com-
other blocks.
pared to their counterparts, driven by a difference with the Table 2 describes the differences in perception and atti-
medical education block. tude between the intervention and control groups. Detail
information about each block can be found in the
The effect of addiction medicine block on the attitude Supplemental Material 2.
toward addiction
There was no main effect of time on attitude, but there was Relationship between illness perceptions and attitude
a main effect of block (F(3, 292) ¼ 3.993, p ¼ .008). This Within the intervention group, several perception subscales
result indicates a more positive attitude in the addiction correlated with the attitude scale. Multicollinearity between
medicine block, compared to others. There was a significant the predictors (the attitude scale before the training, the per-
interaction between time and block for attitude (F(3, ception subscales before and after the training) was not a
292) ¼ 11.13, p < .001), indicating different attitude develop- problem because the VIF scores were below 10 and toler-
ment between blocks. In the intervention group, attitude ance was above 0.2.40 Pearson’s correlation analysis showed
scores increased during the block (baseline: mean ¼ 44.74, only weak to moderate correlations (r < 0.5) between predic-
SD ¼ 6.87; follow-up: mean ¼ 47.16, SD ¼ 7.05), which was tors, except for Consequences baseline and Demoralization
not the case in the control condition. In two control blocks baseline (r ¼ 0.51, p < .001).
(healthcare entrepreneurship and palliative care) the attitude The stepwise regression model revealed three predictors
decreased significantly (baseline: mean ¼ 44.34, SD ¼ 6.00; of post-training attitude: baseline attitude, baseline percep-
follow-up: mean ¼ 41.92, SD ¼ 6.55 and baseline: mean- tion of emotional representation and post-training illness
¼ 44.94, SD ¼ 5.77; follow up: mean ¼ 42.59, SD ¼ 5.35, coherence. The attitude at baseline accounted for 28.9%,
respectively). There was no change in the Medical Education while emotional representation at baseline and illness coher-
block (baseline: mean ¼ 44.89, SD ¼ 6.88; follow-up: mean- ence at follow up accounted for 13.5 and 3.4% of the vari-
¼ 44.94, SD ¼ 7.83). ance in post-training attitude, respectively. The total
 SUBSTANCE ABUSE 5

Table 2. The effect of addiction medicine training on the attitude and perceptions.
Intervention (addiction medicine) Control
(n ¼ 86) (n ¼ 210)
Baseline Follow-up Baseline Follow-up
mean (SD) mean (SD) mean (SD) mean (SD) F(3, 292) pa
Attitude 44.74 (6.87) 47.16 (7.05) 44.70 (6.10) 42.85 (6.51) 11.131 .000b
(MCRS, score: 11–66)
Perception (IPQ-A, score: 1–5)
Emotional representations 3.83 (0.67) 3.96 (0.61) 3.79 (0.68) 3.90 (0.63) 5.277 .001
Demoralization 1.99 (0.49) 1.87 (0.46) 2.13 (0.51) 2.24 (0.59) 10.231 .000c
Illness coherence 3.04 (0.53) 3.81 (0.51) 3.16 (0.59) 3.14 (0.61) 36.456 .000c
Consequences 4.21 (0.49) 4.32 (0.40) 4.22 (0.52) 4.18 (0.51) 3.899 .009
Timeline chronic 3.29 (0.59) 3.57 (0.68) 3.34 (0.62) 3.40 (0.68) 4.662 .003
Patient control 3.93 (0.50) 4.02 (0.51) 3.94 (0.54) 3.89 (0.53) 1.173 .320
Timeline cyclical 3.27 (0.52) 3.29 (0.56) 3.35 (0.53) 3.48 (0.56) 1.838 .140b
Treatment control 3.48 (0.56) 3.55 (0.58) 3.58 (0.60) 3.61 (0.65) 1.496 .216
Attribution (IPQ-A, score: 1–5)
Risk factors 2.75 (0.56) 2.75 (0.68) 2.86 (0.69) 2.83 (0.65) 4.505 .004
Smoking-alcohol 4.11 (0.59) 4.23 (0.61) 4.09 (0.59) 4.06 (0.58) 1.037 .376
Psychological attributions 4.31 (0.52) 4.42 (0.44) 4.26 (0.48) 4.22 (0.48) 6.790 .000b
Overwork 3.67 (0.93) 4.04 (0.68) 3.80 (0.87) 3.71 (0.87) 4.819 .003
a
Effect of the interaction between time and block.
b
Significant difference between addiction block and other blocks (Helmert contrast) (p < .05).
c
Significant difference between addiction block and other blocks (Helmert contrast) (p < .01).
The total attitude score of the Medical Condition Regards Scale. The score of perception scales of Illness Perception Questionnaire Addiction Version. The
score of attribution scales of Illness Perception Questionnaire Addiction Version.

Table 3. Linear model of predictors of the attitude after addiction medi- control group. Finally, we found that students’ attitudes
cine training. before training and illness perceptions, namely emotional
Ba SE bb pc representation (baseline) and illness coherence (follow up),
Step 1 were associated with attitude after training.
Constant 22.49 4.27 <.001
Attitude baseline (MCRS) 0.55 0.09 0.54 <.001 The attitude improvement observed in the intervention
Step 2 group and the declining attitude in the control group sug-
Constant 13.38 4.38 <.01
Attitude baseline (MCRS) 0.40 0.09 0.39 <.001
gests that addiction medicine education contributed to stu-
Emotional representation baseline (IPQ-A) 4.18 0.95 0.40 <.001 dents’ attitude development. These findings are in line with
Step 3 previous studies showing attitude improvement after an
Constant 6.63 5.21 .21
Attitude baseline (MCRS) 0.35 0.09 0.34 <.001 hour of online module,41 a 15-h course in Screening Brief
Emotional representation baseline (IPQ-A) 3.82 0.94 0.36 <.001 Intervention and Referral to Treatment (SBIRT),42 a 1-week
Illness coherence follow up (IPQ-A) 2.72 1.20 0.20 .03 summer school for medical students,43 a 1-week rotation in
Baseline: the measurement prior to addiction medicine training. Follow up: addiction treatment care during psychiatry rotations,27 and a
the measurement after addiction medicine training.
The best model to explain the attitude after the training consisted of 3-week addiction medicine learning module.26 However,
a
three predictors. these studies lack pre and post measurements,43 a compre-
Unstandardized coefficient: correlation coefficient between predictor and out- hensive addiction medicine curriculum,27 or a control condi-
come variable. bStandardized coefficient: correlation coefficient in standard
deviation unit. cSignificance of correlation between predictor and out- tion.26,41,42 Considering such limitations, our study used a
come variable. quasi-experimental design to evaluate a comprehensive
5-week education on addiction medicine by comparing pre-
and post-measurement results between intervention and
control group. By doing this, we improved the method and
variance explained by all three predictors was 45.8%.
confirmed the results of previous studies. In fact, our find-
Table 3 describes the results of the linear regression model.
ings closely align with findings from Christison, et al., where
the attitude score of students after an addiction rotation
Discussion went up from 44 to 47, which is in the same range as in our
study.27 Therefore, it can be speculated that different types
In this study, we evaluated the effect of addiction medicine of addiction medicine education can be beneficial for stu-
training (intervention) on medical students’ attitudes toward dents’ attitude development.
patients with addiction and their illness perceptions of The observed improvement in attitude toward addiction
addiction. We observed that after the intervention, students’ after training is highly relevant, considering the negative
attitudes improved significantly, compared to that of those attitudes and stigma that addicted patients often receive
in the control condition. In addition, participants in the from medical doctors.10,12 Our findings underscore the
intervention group also developed stronger beliefs that they importance of appropriate addiction medicine training dur-
have a coherent understanding of addiction, attributed ing medical school. Given the enormous addiction-related
addiction more to psychological factors, and developed less health burden, all doctors should receive adequate training,
demoralized views toward addiction, as compared to the because the development of more positive attitudes among
6 A. P. AYU ET AL.

medical students might contribute to better care for these more positive attitude during training, as did those who
patients.10,22,44 developed a better understanding. The instruments used in
Our findings further suggest that addiction medicine this study intend to measure what health professionals think
training may influence perceptions of addiction among med- about the condition addiction (perception) and their actual
ical students, which has never been studied before. attitude toward patients with those conditions (attitude). It
Specifically, medical students in the intervention group has been suggested that how people perceive addiction (per-
expressed a less demoralized view toward addiction, devel- ception) is associated with their attitude toward patients in
oped stronger beliefs that they have a coherent understand- clinical practice (attitudes).18 As such, this study aimed at
ing of addiction, and attributed addiction to psychological gaining a better understanding of the relationship between
factors, as compared to the control group. This suggests that illness perception and attitude in the context of education
addiction medicine training might be an effective way of that provides knowledge about addiction for medical stu-
developing perceptions of addiction in medical students. dents. Although these are quite abstract concepts, which are
The Demoralization subscale refers to demoralization not easy to delineate sharply, this study found two percep-
among the respondents. It is associated with feelings of tions to mainly influence the attitude development of stu-
helplessness and hopelessness about the condition, meaning dents who received addiction training. Though we cannot
that nothing can be done, either by the patient or treatment, infer any causal conclusions, our observations support the
that can improve the patient’s condition.37 The 5-week idea that the perception that a condition is emotionally
addiction medicine block seems to stimulate students’ sense stressful and a better understanding of the condition might
of hope toward the patients’ condition. More coherent contribute to more positive attitudes. It has been suggested
understanding seems to fit with gaining knowledge and that interaction with patients with addiction may be particu-
skills related to addiction medicine and the development of larly relevant to improve understanding of addiction, which
a professional frame of mind. Though illness perceptions subsequently improves the attitude.26,47
cannot be considered right or wrong; positive or negative,45 The regression model accounted for nearly 46% of the
it is tempting to speculate that particularly low demoraliza- variance in attitude after the training. This means that other
tion and perceived better understanding could contribute to
factors are also important for attitude toward patients with
more positive attitudes toward addiction.
addiction. For example, we did not assess knowledge, skills,
Though some perceptions assessed with the IPQ-A may
personal experience with substance use, stigma, the teachers’
not align with current scientific knowledge (e.g., addiction
values concerning addiction, and the influence of drug pol-
cannot be cured), these perceptions might still be present,
icy.26,48–50 Indonesia has a strict policy concerning substance
also among health professionals. Especially for a stigmatized
use and applies the death penalty for drug traffickers.51 Such
condition as addiction, such nonscientific perceptions might
a strong punitive policy might influence how society, includ-
be highly relevant for attitude. For example, a clinician who
ing the medical community, perceives drug use.52 Moreover,
holds demoralized perception of addiction, and the percep-
although the majority of people living in Indonesia are
tion that treatment cannot control addiction is less likely to
Muslim, this study was conducted in a catholic university,
motivate a patient with addiction for treatment, despite a
large body of evidence for treatment effectiveness.44 with a track record in teaching and research in addiction
The attitude before training emerged as an important fac- topic. We did not evaluate the university’s value or policy
tor for the development of attitude among addiction medi- concerning substance use. These might be relevant factors to
cine block students after the training. In other words, the take into account in future studies. Furthermore, teachers’
more positive attitude at baseline, the more positive the atti- attitudes toward conditions are referred to as the “hidden
tude becomes at follow-up. It is important to point out here curriculum” that can influence attitude development among
that we analyzed the predictors of attitude after training medical students.53
only within the intervention group. Therefore, the fact that The results of our study should be interpreted in light of
all groups had similar baseline attitudes is therefore not rele- its strengths and weaknesses. First, attitude toward addiction
vant in the context of this finding. was assessed by self-report. One study compared the effect
One previous study showed that attitude only improved of a 6-month addiction medicine training in general
among general practitioners who had a positive attitude practitioners on self-reported attitude and self-reported
before an addiction medicine training and deteriorated behavior.54 While there was a significant improvement in
among those who already had a negative attitude.46 Though self-reported attitude, this was not the case for the self-
it might not be feasible to tailor addiction medicine training reported behavior. Therefore, it remains to be seen whether
to specific individual training needs or individual attitudes, the observed improvement in self-reported attitude actually
it might be relevant to keep this in mind when developing reflects improved behavior in real-life. It is also not known
addiction medicine curricula or when teaching addiction what numeric difference should be considered clinically
medicine to medical students. meaningful, and whether attitude differences indexed by this
Of all perceptions, emotional representation (baseline) instrument translate into actual differences in clinical behav-
and illness coherence (follow up) were associated with atti- ior.14 Qualitative methods such as semi-structured interviews
tude after the training. In other words, students who per- might be able to capture more aspects of illness perceptions
ceived addiction to be emotionally stressful developed a and attitude than self-report questionnaires.
 SUBSTANCE ABUSE 7

Second, the post-measurement of the training effect dir- Margot Van Der Ven contributed in the collection of data,
ectly followed the training. One previous study reported per- interpretation of the result, and writing the manuscript.
sisting positive effects of addiction medicine training, Eva Suryani contributed in the collection of data and
provided in the second year, on medical students’ attitude revision of the manuscript.
up to their graduation four years later.55 More studies with Natalia Puspadewi contributed in the interpretation of
long-term follow-up are needed to evaluate the sustainability the results and revision of the manuscript.
of the observed effects of addiction medicine training Satya Joewana contributed in research conception and
over time. design, interpretation of the results and revision of
Third, we carried out this study at the only medical school the manuscript.
in Indonesia with specific comprehensive, 5-week addiction Elisabeth Rukmini contributed in research conception
medicine training. This limits the generalizability of the and design, interpretation of the result, and revision of
results. Nevertheless, since this study is the first on this topic the manuscript.
from South East Asia, it offers new insight concerning addic- Cor De Jong contributed in research conception and
tion medicine education, particularly for this region. design, interpretation of the result, and revision of
Additionally, future studies might also explore perceptions the manuscript.
and attitudes toward addiction in other professionals. Arnt Schellekens contributed in research conception and
Fourth, the students chose their preferred block, although design, analysis of data, interpretation of the result, and
some students were transferred to the addiction medicine writing and revision of the manuscript.
block because of the limited availability at their preferred
block. It seems that the attitude prior to training did not influ- Funding
ence students’ decision to choose a certain block since we did
not find a significant difference in attitude between the inter- Author Astri Parawita Ayu received a grant to perform this research
from the Directorate General of Resources for Research, Technology
vention and control group at baseline. Nevertheless, it is also and Higher Education, Ministry of Research, Technology, and Higher
important to consider that students’ preference of block might Education of the Republic of Indonesia [94.19/E4.4/2014]. The funder
be motivated by other aspects of attitude that cannot be has no role in study design, in data collection, analysis, and interpret-
addressed by the MCRS. In addition, other factors, such as ation, in the writing of the report, and in the decision to submit the
culture or the activities proposed by the block might also article for publication.
encourage students to enroll in a certain block. Despite the
fact that we did not find any significant differences in attitude
at baseline, any selection bias cannot be fully ruled out. ORCID
Astri Parawita Ayu http://orcid.org/0000-0003-0979-1593
Conclusions Cor de Jong http://orcid.org/0000-0003-1824-7303
Arnt Schellekens http://orcid.org/0000-0002-7715-5209
In conclusion, our findings strengthen the evidence of the
effectiveness of addiction medicine training in improving
medical students’ attitudes toward patients with addiction. References
Moreover, we also showed that illness perception can be [1] Smith DE. The process addictions and the new ASAM defin-
influenced by education, which may influence attitude devel- ition of addiction. J Psychoact Drugs. 2012;44(1):1–4.
[2] UNODC. World Drug Report 2016. Vienna, Austria: United
opment. These findings show that perception and attitude
Nations Office on Drugs and Crime; 2016.
are relevant in the context of addiction medicine training. [3] Degenhardt L, Whiteford HA, Ferrari AJ, et al. Global burden
Though we did not provide evidence that attention for atti- of disease attributable to illicit drug use and dependence: find-
tude and perception as part of the curriculum are the effect- ings from the Global Burden of Disease Study 2010. The
ive ingredients of the curriculum, we suggest that addiction Lancet. 2013;382(9904):1564–1574.
[4] Bilano V, Gilmour S, Moffiet T, et al. Global trends and projec-
medicine training for medical students should not solely
tions for tobacco use, 1990–2025: an analysis of smoking indi-
focus on knowledge and skills development, but should also cators from the WHO Comprehensive Information Systems for
incorporate aspects of attitude and personal views on addic- Tobacco Control. The Lancet. 2015;385(9972):966–976.
tion. Providing such comprehensive addiction medicine [5] Reitsma MB, Fullman N, Ng M, et al. Smoking prevalence and
training at undergraduate medical school, including reflec- attributable disease burden in 195 countries and territories,
tion and attitude development, might contribute to a better 1990–2015: a systematic analysis from the Global Burden of
Disease Study 2015. The Lancet. 2017;389(10082):1885–1906.
quality of medical care for patients with addiction in [6] Lai HMX, Cleary M, Sitharthan T, Hunt GE. Prevalence of
the future. comorbid substance use, anxiety and mood disorders in epi-
demiological surveys, 1990 - 2014: a systematic review and
meta-analysis. Drug Alcohol Depend. 2015;154:1–13.
Author contributions [7] Bhargava S, Arora RR. Cocaine and cardiovascular complica-
tions. Am J Therap. 2011;18(4):e95–e100.
Astri Parawita Ayu contributed in research conception and [8] Megarbane B, Chevillard L. The large spectrum of pulmonary
design, analysis of data, interpretation of the result, and complications following illicit drug use: features and mecha-
writing and revision of the manuscript. nisms. Chemico Biol Interact. 2013;206(3):444–451.
8 A. P. AYU ET AL.

[9] Chorlton E, Smith IC. Understanding how people with mental [29] Korszun A, Dinos S, Ahmed K, Bhui K. Medical student atti-
health difficulties experience substance use. Subst Use Misuse. tudes about mental illness: does medical-school education
2016;51(3):318–329. reduce stigma? Acad Psychiatry. 2012;36(3):197.
[10] Gilchrist G, Moskalewicz J, Slezakova S, et al. Staff regard [30] Boyle MJ, Williams B, Brown T, et al. Attitudes of undergradu-
towards working with substance users: a European multi-centre ate health science students towards patients with intellectual
study. Addiction. 2011;106(6):1114–1125. disability, substance abuse, and acute mental illness: a cross-sec-
[11] Fernando SM, Deane FP, McLeod HJ. Sri Lankan doctors’ and tional study. BMC Med Educ. 2010;10(1):1–8.
medical undergraduates’ attitudes towards mental illness. Soc [31] Brown T, Williams B, Boyle M, et al. Levels of empathy in
Psychiat Epidemiol. 2010;45(7):733–739. undergraduate occupational therapy students. Occup Ther Int.
[12] VanBoekel LC, Brouwers EPM, VanWeeghel J, Garretsen H. 2010;17(3):135–141.
Stigma among health professionals towards patients with sub- [32] McKenna L, Boyle M, Brown T, et al. Level of empathy in
stance use disorders and its consequences for healthcare deliv- undergraduate midwifery students: an Australian cross-sectional
ery: systematic review. Drug Alcohol Depend. 2013;131:23–25. study. Women and Birth. 2011;24(2):80–84.
[13] Russell C, Davies JB, Hunter SC. Predictors of addiction treat- [33] Madhan B, Gayathri H, Garhnayak L, Naik ES. Dental students’
ment providers’ beliefs in the disease and choice models of regard for patients from often-stigmatized populations: findings
addiction. J Subst Abuse Treat. 2011;40(2):150–164. from an Indian dental school. J Dent Educ. 2012;76(2):210–217.
[14] Christison GW, Haviland MG, Riggs ML. The medical condi- [34] Williams B, Boyle M, Fielder C. Empathetic attitudes of under-
tion regard scale: measuring reactions to diagnoses. Acad Med. graduate paramedic and nursing students towards four medical
2002;77(3):257–262. conditions: a three-year longitudinal study. Nurse Education
[15] Brener L, Hippel WV, Hippel CV, Resnick I, Treloar C. Today. 2015;35(2):e14–e18.
Perceptions of discriminatory treatment by staff as predictors of [35] Moss-Morris R, Weinman J, Petrie K, Horne R, Cameron L,
drug treatment completion: utility of a mixed methods Buick D. The revised illness perception questionnaire (IPQ-R).
approach. Drug Alcohol Rev. 2010;29(5):491–497. Psychology & Health. 2002;17(1):1–16.
[16] VanBoekel LC, Brouwers EPM, VanWeeghel J, Garretsen H. [36] Weinman J, Petrie KJ. Illness perceptions: a new paradigm for
Experienced and anticipated discrimination reported by individ- psychosomatics? J Psychosom Res. 1997;42(2):113–116.
[37] Ayu AP, Dijkstra B, Golbach M, DeJong C, Schellekens A.
uals in treatment for substance use disorders within the
Good psychometric properties of the addiction version of the
Netherlands. Health Social Care in the Commun. 2015;24(5):
revised illness perception questionnaire for health care profes-
e23–e33.
sionals. PLoS One. 2016;11(11):e0164262.
[17] VanBoekel LC, Brouwers EPM, VanWeeghel J, Garretsen H.
[38] Weinman J, Petrie KJ, Moss-Morris R, Horne R. The illness
Inequalities in healthcare provision for individuals with sub-
perception questionnaire: a new method for assessing the cogni-
stance use disorders: perspectives from healthcare professionals
tive representation of illness. Psychol Health. 1996;11(3):
and clients. J Subst Use. 2016;21:133–140.
431–455.
[18] Brener L, Hippel WV, Kippax S, Preacher KJ. The role of phys-
[39] IBM. SPSS statistics: what’s new. New York, NY: IBM
ician and nurse attitudes in the health care of injecting drug
Corporation; 2017.
users. Subst Use Misuse. 2010;45(7–8):1007–1018. [40] Field A. Discovering Statistic Using IBM SPSS Statistics. 4th ed.
[19] Husak DN. The moral relevance of addiction. Subst Use Misuse.
London, UK: SAGE publications Ltd.; 2013.
2004;39(3):399–436. [41] Barone EJ, Huggett KN, Lofgreen AS. Investigation of students’
[20] Alexander BK. Addiction: the urgent need for a paradigm shift.
attitudes about patients with substance use disorders before and
Subst Use Misuse. 2012;47:1475–1482. after completing an online curricular module. Ann Behav Sci
[21] Oliva EM, Maisel NC, Gordon AJ, Harris A. Barriers to use of
Med Educ. 2011;17(1):10–13.
pharmacotherapy for addiction disorders and how to overcome [42] Koyi MB, Nelliot A, MacKinnon D, et al. Change in
them. Curr Psychiatry Rep. 2011;13(5):374–381. medical student attitudes toward patients with substance use
[22] Ayu AP, Schellekens AFA, Iskandar S, Pinxten L, DeJong C. disorders after course exposure. Acad Psychiatry. 2018;42(2):
Effectiveness and organization of addiction medicine training 283–287.
across the globe. Eur Addict Res. 2015;21(5):223–239. [43] Barron R, Frank E, Gitlow S. Evaluation of an experiential cur-
[23] Kothari D, Gourevitch MN, Lee JD, et al. Undergraduate med- riculum for addiction education among medical students. J
ical education in substance abuse: a review of the quality of the Addict Med. 2012;6(2):131–136.
literature. Acad Med. 2011;86(1):98–112. [44] Wood E, Samet JH, Volkow ND. Physician education in addic-
[24] Spangler JG, George G, Foley KL, Crandall SJ. Tobacco inter- tion medicine. JAMA. 2013;310(16):1673–1674.
vention training: current efforts and gaps in US medical school. [45] Petrie KJ, Weinman J. Why illness perceptions matter. Clin
J Am Med Assoc. 2002;288(9):1102–1109. Med. 2006;6(6):536–539.
[25] Kahan M, Wilson L, Midmer D, Borsoi D, Martin D. [46] Anderson P, Kaner E, Wutzke S, et al. Attitudes and managing
Randomized controlled trial on the effects of a skills-based alcohol problems in general practice: an interaction analysis
workshop on medical students’ management of problem drink- based on findings from a WHO collaborative study. Alcohol &
ing and alcohol dependence. Subst Abuse. 2003;24(1):5–16. Alcoholism. 2004;39:351–356.
[26] Silins E, Conigrave KM, Rakvin C, Dobbins T, Curry K. The [47] Ding L, Landon BE, Wilson IB, Wong MD, Shapiro MF, Cleary
influence of structured education and clinical experinece on the PD. Predictors and Consequences of Negative Physician
attitudes of medical students towards substance misusers. Drug Attitudes toward HIV-Infected Injection Drug Users. Arch
& Alcohol Revs. 2007;26(2):191–200. Intern Med. 2005;165:618–623.
[27] Christison GW, Haviland MG. Requiring a one-week addiction [48] VanBoekel LC, Brouwers EPM, VanWeeghel J, Garretsen H.
treatment experience in a six-week psychiatry clerkship: effects Healthcare professionals’ regard towards working with patients
on attitudes toward substance-abusing patients. Teach Learn with substance use disorders: comparison of primary care, gen-
Med. 2003;15(2):93–97. eral psychiatry and specialist addiction services. Drug Alcohol
[28] Rudnick A. Attitudes of preclinical medical students towards Depend. 2014;134:92–98.
psychiatric patients before and after an early clinical experience. [49] Raistrick D, Russell D, Tober G, Tindale A. A survey of sub-
Can Med Educ J. 2011;2(1):e11–e15. stance use by health care professionals and their attitudes to
 SUBSTANCE ABUSE 9

substance misuse patients (NHS Staff Survey). J Subst Use. [53] Birden H, Glass N, Wilson I, Harrison M, Usherwood T,
2008;13(1):57–69. Nass D. Defining professionalism in medical education: a sys-
[50] Lawrence RE, Rasinski KA, Yoon JD, Curlin FA. Physicians’ tematic review. Med Teach. 2014;36(1):47–61.
beliefs about the nature of addiction: a survey of primary care [54] Strang J, Hunt C, Gerada C, Marsden J. What difference does
physicians and psychiatrists. Am J Addict. 2013;22(3):255–260. training make? A randomized trial with waiting-list control of
[51] Lai G, Asmin F, Birgin R. Drug policy in Indonesia. general practitioners seeking advanced training in drug misuse.
International Drug Policy Consortium. 2013:1–15. Addiction. 2007;102(10):1637–1647.
[52] Ayu AP, Iskandar S, Siste K, DeJong C, Schellekens A. [55] Gopalan R, Santora P, Stokes EJ, Moore RD, Levine DM.
Addiction training for health professionals as an antidote to the Evaluation of a model curriculum on substance abuse at the
addiction health burden in Indonesia. Addiction. 2016;111(8): Johns Hopkins University school of medicine. Acad Med. 1992;
1498–1499. 67(4):260–266.