Substance Use, Substance Use Disorders, and Co-Occurring Psychiatric Disorders in Recently Incarcerated Men: A Comparison with the General Population

Abstract

Introduction
The rates of alcohol and illegal drug use and the prevalence of alcohol and illegal drug use disorders (AUD and DUD) are high in prison populations, particularly in men entering jail. However, these rates have never been exhaustively assessed and compared to those of the general population in France.
Methods
We based our research on two surveys, conducted in the same French region, which included a total of 630 men entering jail and 5,793 men recruited from the general population. We used the Mini International Neuropsychiatric Interview to assess alcohol and drug use, AUD, DUD, as well as co-occurring psychiatric disorders and we examined differences in prevalence rates between the two populations. Logistic regression models were performed to (i) identify the factors associated with AUD and DUD, (ii) test whether the interaction between admission to jail and the presence of AUD, DUD, or both is linked to the presence of at least one co-occurring psychiatric disorder.
Results
Compared to the general population sample, the prevalence of AUD (33.8% vs 8.7%, p<0.001) and DUD (at least one type of drug: 28.7% vs 5.0%, p<0.001; cannabis: 24.0% vs 4.7%, p<0.001; opioids: 6.8% vs 0.4%, p<0.001; stimulants: 5.2% vs 0.8%, p<0.001) were significantly higher in the jail population sample, as well as the rates of past-year use of various substances (alcohol: 62.1% vs 56.4%, p=0.007; at least one type of illegal drug: 50.0% vs 14.4%, p<0.001; cannabis: 45.6% vs 13.9%, opioids: 9.4% vs 0.7%; stimulants: 8.6% vs 1.9%). Admission to jail was associated with a higher risk of AUD (aOR=3.80, 95% CI: 2.89-5.01, P<0.001) or DUD (aOR=4.25, 95% CI: 3.10-5.84, P<0.001). History of trauma was also associated with both AUD (aOR=1.81, 95% CI: 1.53-2.14, P<0.001) and DUD (aOR=2.15, 95% CI: 1.74-2.65, P<0.001) whereas history of migration was only associated with DUD (aOR=1.38, 95% CI: 1.12-1.71, P=0.003). AUD and DUDs were more strongly associated with co-occurring psychiatric disorders in incarcerated men than in the general population. Among individuals with AUD, DUD or both, co-occurring anxiety and mood disorders were particularly more frequent in jail than in the general population.
Discussion/Conclusion
As in most countries, AUD and DUD are highly prevalent among men entering jail in France. Our results also suggest that incarceration constitutes an independent vulnerability factor for a dual disorder, which supports a systematic assessment and treatment of psychiatric disorders in men entering jail and diagnosed with an AUD or DUD.

Full text

1
Substance*use,*substance*use*disorders,*and*co-occurring*psychiatric*
disorders*in*recently*incarcerated*men:*a*comparison*with*the*
general*population**
Thomas Foveta,b, Marielle Watheletb,c, Massil Benbouriched,e, Imane Benradiaf,g,
Jean-LucRoelandtf,g, Pierre Thomasa,c, Fabien D’Hondta,b, Benjamin Rollandh,i
a Univ. Lille, Inserm, CHU Lille, U1172 - Lille Neuroscience & Cognition, F-59000 Lille, France
b Centre national de ressources et de résilience Lille-Paris (CN2R), F-59000 Lille, France
c Fédération régionale de recherche en psychiatrie et santé mentale, Hauts-de-France, France
d Univ. Lille, ULR 4072 - PSITEC - Psychologie : Interactions Temps Émotions Cognition, F-59000 Lille,
France
e Centre de Recherche, Institut National de Psychiatrie Légale Philippe-Pinel, Montréal, Québec
f EPSM Lille Métropole, Centre Collaborateur de l’Organisation Mondiale de la Santé pour la recherche et la
formation en santé mentale, Lille, France.
g ECEVE, UMRS 1123, Université Paris Diderot, Sorbonne Paris Cité, INSERM, Paris, France
h Service Universitaire d’Addictologie de Lyon (SUAL), Hospices Civils de Lyon, CH Le Vinatier, France
i Centre de Recherche en Neurosciences de Lyon (CRNL), Équipe PsyR², INSERM U1028, CNRS UMR5292,
Université Claude Bernard Lyon 1, Université Jean Monnet Saint Etienne, Lyon, France
Short Title: Substance use patterns among men entering jail in France
Corresponding author :
Thomas Fovet
Unité hospitalière spécialement aménagée
Chemin du Bois de l'Hôpital
59113 SECLIN, FRANCE.
Tel : 0361763002
E-mail: Thomas.fovet@chru-lille.fr

2
Number of tables: 3
Number of figures: 0
Word count: 3728 words
References: 35
Supplementary material: 1 Supplementary Table/1 Supplementary Figure
Keywords: Jail; France; Substance use disorder; Misuse; Comorbidity.
Acknowledgement
The authors thank Emeline BRENIAUX for her help in editing the manuscript.
Statement of Ethics
The authors assert that all procedures contributing to this work comply with the ethical
standards of the relevant national and institutional committees on human experimentation and
with the Helsinki Declaration of 1975, as revised in 2008.
Legal authorization was obtained by the “Comité de Protection des Personnes” (CPP) –
authorization number IDRCB 2012 A0144835, the “Commission Nationale Informatique et
Liberté” (CNIL) – authorization number MMS/VCS/AR152838, and the “Agence nationale de
sécurité du médicament et des produits de santé” (ANSM) – authorization number 130500B-
31.
All interviewees provided written informed consent.
Conflict of Interest Statement
The authors declare no conflicts of interest.

3
Funding Sources
The Mental Health in the Prison Population survey was supported by the Agence Régionale de
Santé des Hauts de France.
Author Contributions
Thomas Fovet, Marielle Wathelet, Fabien D’Hondt, and Benjamin Rolland participated in the
conception and design of the study; Imane Benradia, Jean-Luc Roelandt, and Pierre Thomas
participated in the acquisition of data; Marielle Wathelet performed the analyses; Thomas Fovet
and Benjamin Rolland wrote the first draft of the manuscript. Thomas Fovet, Marielle Wathelet,
Massil Benbouriche, Imane Benradia, Jean-Luc Roelandt, Pierre Thomas, Fabien D’Hondt, and
Benjamin Rolland participated in the writing and revision of the successive drafts of the
manuscript and approved the final version.
Data Availability Statement
The data on which this manuscript is based are not publicly available. However, data from
Mental Health-Prison Population (MH-Prison Population) are available upon request, under the
supervision of the principal investigator (PI) of the study. Access to data needs to be subjected
to a formal data sharing agreement to ensure that any other researcher accessing the data follows
appropriate ethical standards. The PI of MH-Prison Population (Pr. Pierre Thomas) can be
contacted (pierre.thomas@chru-lille.fr) at all times to request data: researchers can submit a
research plan, describing its background, research questions, variables to be used in the
analyses, and an outline of the analyses. If such a request is approved, a written agreement will
be signed stating that the data will only be used for addressing the agreed research questions,
and not for other purposes.

4
Abstract
Introduction
The rates of alcohol and illegal drug use and the prevalence of alcohol and illegal drug use disorders
(AUD and DUD) are high in prison populations, particularly in men entering jail. However, these rates
have never been exhaustively assessed and compared to those of the general population in France.
Methods
We based our research on two surveys, conducted in the same French region, which included a total of
630 men entering jail and 5,793 men recruited from the general population. We used the Mini
International Neuropsychiatric Interview to assess alcohol and drug use, AUD, DUD, as well as co-
occurring psychiatric disorders and we examined differences in prevalence rates between the two
populations. Logistic regression models were performed to (i) identify the factors associated with AUD
and DUD, (ii) test whether the interaction between admission to jail and the presence of AUD, DUD, or
both is linked to the presence of at least one co-occurring psychiatric disorder.
Results
Compared to the general population sample, the prevalence of AUD (33.8% vs 8.7%, p<0.001) and
DUD (at least one type of drug: 28.7% vs 5.0%, p<0.001; cannabis: 24.0% vs 4.7%, p<0.001; opioids:
6.8% vs 0.4%, p<0.001; stimulants: 5.2% vs 0.8%, p<0.001) were significantly higher in the jail
population sample, as well as the rates of past-year use of various substances (alcohol: 62.1% vs 56.4%,
p=0.007; at least one type of illegal drug: 50.0% vs 14.4%, p<0.001; cannabis: 45.6% vs 13.9%, opioids:
9.4% vs 0.7%; stimulants: 8.6% vs 1.9%). Admission to jail was associated with a higher risk of AUD
(aOR=3.80, 95% CI: 2.89-5.01, P<0.001) or DUD (aOR=4.25, 95% CI: 3.10-5.84, P<0.001). History of
trauma was also associated with both AUD (aOR=1.81, 95% CI: 1.53-2.14, P<0.001) and DUD
(aOR=2.15, 95% CI: 1.74-2.65, P<0.001) whereas history of migration was only associated with DUD
(aOR=1.38, 95% CI: 1.12-1.71, P=0.003). AUD and DUDs were more strongly associated with co-
occurring psychiatric disorders in incarcerated men than in the general population. Among individuals
with AUD, DUD or both, co-occurring anxiety and mood disorders were particularly more frequent in
jail than in the general population.
Discussion/Conclusion
As in most countries, AUD and DUD are highly prevalent among men entering jail in France. Our results
also suggest that incarceration constitutes an independent vulnerability factor for a dual disorder, which
supports a systematic assessment and treatment of psychiatric disorders in men entering jail and
diagnosed with an AUD or DUD.

5
Introduction
Eleven million people are currently incarcerated around the world [1]. Among the many
health issues affecting the prison population, substance use and substance use disorders (SUDs)
occupy a prominent place [2]. A recent meta-analysis of 24 studies including a total of 18,388
people in prison across 10 countries identified a pooled prevalence estimate of alcohol use
disorder (AUD) of 26% (95% confidence interval [CI] = 23-30) in men and 20% (95%
confidence interval [CI] = 16-24) in women. In the same study, the pooled prevalence estimate
of illegal drug use disorders (DUDs) was 30% (95% CI = 22-38) and 51% (95% CI = 24-58) in
men and women in prison, respectively [3]. The implications for this population are substantial:
SUDs are associated with many specific adverse health outcomes in people in prison, including
suicide in custody [4,5], early death after release [6,7], and judicial consequences such as
repeated offending [8] or perpetrating violence during incarceration [9].
Despite the high prevalence of SUDs among people in prison and the many related
negative outcomes, there are still many shortcomings in the management of these disorders in
prisons and correction facilities [10,11]. Optimizing care requires several interventions at
different stages of incarceration, including identifying and treating withdrawal on admission to
prison, appropriately prescribing SUD treatment (e.g., opioid agonist treatments), providing
people with tailored psychotherapy throughout the period of incarceration, and planning the
release to avoid treatment interruptions. Improving the quality of care at all these stages requires
a good knowledge of the clinical characteristics of people suffering from SUDs in prison in
order to adapt the provision of care (often limited in prison) more accurately to health needs.
Importantly, many studies have shown that the co-occurrence of SUDs and psychiatric
disorders is frequent in the prison population [12–15]. This suggests that people who are
incarcerated may have specific mental health needs and may need tailored treatments [2].

6
Thorough knowledge of the most frequently used substances is also essential for appropriate
management.
The epidemiology of SUDs in prisons and correction facilities is considerably different
from country to country. In males, the prevalences in individual studies included in the Fazel et
al. meta-analysis (2017) ranged from 16 to 51% and from 10 to 61% for AUDs and DUDs
respectively. The rates of heterogeneity between the included studies were very high in this
meta-analysis, which could be explained by the widely varying community baseline rates of
SUDs in each country. Studies using comparable measures in both the prison environment and
the community are thus urgently needed. In France, Mental Health in the Prison Population
(MH-Prison Population) and Mental Health in the General Population (MH-General
Population) are two surveys conducted using the same methodology in both prison and general
populations, respectively. Their combined analysis offers the opportunity to compare both
populations with respect to past-year substance use, SUDs, and co-occurring psychiatric
disorders.
The main objective of this study was to assess and compare the rates of past-year
substance use, as well as the prevalence of AUD and DUD, in samples of men from the general
population and men entering jail. Secondary objectives included determining the
sociodemographic factors correlated with AUD and DUD and testing the hypothesis that there
is an interaction between admission to jail and the presence of a DUD, an AUD, or both on the
presence of co-occurring psychiatric disorders.

7
Material and methods
Study population
This study is an explorative secondary analysis of the MH-Prison Population and the
MH-General Population surveys. We focused on the male population since women represent
only a small proportion of the prison population (3.5% of the participants who responded to the
MH-Prison Population [16]), and since significant differences in the mental health status of
incarcerated men and women are well documented [3,17].
Mental Health in the Prison Population survey (MH-Prison Population)
The cross-sectional MH-Prison Population survey, conducted between March 2014 and
April 2017 by the Fédération régionale de recherche en psychiatrie et santé mentale (Regional
Federation for Research in Psychiatry and Mental Health, F2RSMPsy) and the World Health
Organization Collaborating Centre in Mental Health (WHOCC Lille), included 653 men and
women who had recently been committed to the French general population prison system in the
French departements Nord and Pas-de-Calais. A total of 934 subjects were randomly
approached but only 655 were eligible and provided consent to participate in the study
(participation rate: 70.1%). Two participants were excluded because of incomplete data. In
total, 653 participants were interviewed. The present study analyzed the data from all 630 men
interviewed (23 women were excluded).
Legal authorization was obtained by the “XXXX” – authorization number XXXX, the
“XXXX” – authorization number XXXX, and the “XXXX” – authorization number XXXX.
All interviewees provided written informed consent. The data were collected between March
2014 and April 2017. Each participant was interviewed for 45 to 60 minutes within the jail
medical unit, under strict conditions of confidentiality.

8
Participants were randomly approached for inclusion from admission lists of people
committed in eight of the nine jails of the French departments Nord and Pas-de-Calais: Arras,
Douai, Dunkirk, Sequedin, Annoeullin, Longuenesse, Maubeuge, and Valenciennes. In France,
jails are remand centers or detention centers where people whose sentence is shorter than two
years reside. Participants were included in the study if they met the following criteria: (1)
provided informed consent to participate in the survey, (2) spoke French, (3) were aged 18 years
or older, (4) were free of any mental or psychological incapacity to participate due to acute
decompensation of a psychiatric disorder or severe substance withdrawal (i.e., a health
condition requiring emergency hospitalization), and (5) had been incarcerated for less than 72
hours (sentenced or on remand). For each jail, the recruitment days were selected at random.
People with a safeguarding vulnerable adult legal measure were included. For a full description
of the MH-Prison Population survey procedure, see [16].
Mental Health in the General Population survey (MH-General Population)
The cross-sectional MH-General Population survey, conducted by the WHOCC,
interviewed 12,568 subjects in the French department Nord between 2001 and 2008.
Participants were selected from 14 sites (900 subjects per site) using a quota-sampling method.
This method provides a sample of subjects with a sociodemographic profile similar to that of
the general population profile with regard to age, sex, education level, and occupational
category according to census figures from 1999 provided by the French National Institute of
Statistics and Economic Studies (INSEE). The present study analyzed the data from all 5,793
men interviewed. No participant was excluded due to incomplete data.
Participants were included in the study if they met the following criteria: (1) provided
informed consent to participate in the survey, (2) spoke French, (3) were aged 18 years or older,
(4) were residing in the French departments Nord or Pas-de-Calais, and (5) were neither

9
institutionalized nor homeless. A full description of the MH-General Population survey
procedure is available elsewhere [18–22].
Measures
The MH-Prison Population and MH-General Population surveys used the same
methodology to collect (i) clinical measures , (ii) sociodemographic data.
Main outcomes
For each participant, the Mini-International Neuropsychiatric Interview (MINI; French
version 5.0.0), a standardized psychiatric interview, was used to screen for SUDs (AUD and
DUD were considered separately) as defined by the 10th International Classification of
Diseases (ICD-10) [23]. The following SUDs were assessed: (1) AUD (i.e., alcohol abuse [past
12 months, F10.1], or alcohol dependence [past 12 months, F10.2]); (ii) DUD (i.e., substance
abuse [past 12 months, F1X.1], or substance dependence [past 12 months, F1X.2], excluding
AUD).
Past-year substance use was explored using the first question of the MINI: (1) “In the
past 12 months, have you had 3 or more alcoholic drinks within a 3-hour period on 3 or more
occasions?” (an alcoholic drink refers to a standard unit) for alcohol, and (2) “In the past 12
months, did you take any of these drugs more than once?” for illegal drugs.
Substances were grouped as follows: cannabis (i.e., cannabis, hash, marijuana, or
hashish), opioids (i.e., heroin, buprenorphine, methadone, codeine, morphine, opium),
stimulants (i.e., cocaine, ecstasy, amphetamines, crack, Artane, Captagon, Catovit, Ritalin, coca
leaf), psychedelics (i.e., LSD, mescaline), and other substances. Tobacco was not considered.
Co-occurring psychiatric disorders
The following psychiatric disorders were assessed using the MINI: (i) mood disorders,
i.e., manic episodes (lifetime, F30), depressive disorders (current [2 weeks] and recurrent, F32

10
and F33), and dysthymia (current [past 2 years], F34.1); (ii) anxiety disorders, i.e., panic
disorder with or without agoraphobia (current, F41.0 and F40.01), social anxiety disorder
(SAD, current, F40.1), and generalized anxiety disorder (GAD, current [past 6 months] F41.1);
(iii) psychotic syndromes (lifetime, isolated or recurrent, F2[x]), always confirmed by a senior
psychiatrist or psychologist); (iv) posttraumatic stress disorder (PTSD, current, F43.1). Suicide
risk (lifetime) was also evaluated, based on the MINI (low, medium, or high risk).
Good inter-rater and test-retest reliabilities of the MINI have been previously reported.
A good convergent validity relative to the Composite International Diagnostic Interview (CIDI)
and the Structured Clinical Interview for Diagnostic and Statistical Manual (SCID) has also
been described [23–25]. All of the MH-Prison Population interviewers (nurses and
psychologists) were trained by WHOCC experts to conduct the MINI over a 1-day session.
Sociodemographic data
To characterize the samples, sociodemographic data were collected on age in years
(categorized into four bands: 18-24, 25-34, 35-44, and over 44), education level (no education
or primary level, secondary level, university level), marital status (single, married or coresiding
with a partner, separated/divorced/widowed), employment status (employed, unemployed),
monthly income in euros per household per month (categorized into five bands: <€534, €535-
840, €841-1,300, €1,301-2,520, >€2,520) at the time of the interview. History of migration
(from any country) was screened by asking the participant his country of birth, as well as that
of his parents and grandparents (if the participant or one of his parents or grandparents was born
outside France, we rated yes, otherwise no). History of trauma exposure, i.e., having been
exposed to a stressful event or situation of exceptionally threatening or catastrophic nature,
which would likely cause pervasive distress in almost anyone (yes, no) were also assessed.

11
Statistical analyses
Crude rates (and 95% confidence intervals [95%CI]) of past-year substance use, AUD,
and DUD, were measured in the MH-General Population and MH-Prison Population samples.
Age-standardized prevalences (95%CI) in the MH-Prison Population sample were also
calculated using the direct method and considering the MH-General Population sample age
structure as a reference [26]. Prevalences in the two samples were compared using Chi-square
or Fisher tests.
Logistic regression models were performed to identify the factors associated with AUD
and DUD, including the following explanatory variables: age category, education level, marital
status, employment status, monthly income, history of migration, and history of trauma.
Associations were presented as adjusted odds ratios and 95%CI (aOR [95%CI]).
Finally, we tested the hypothesis that there is an interaction between admission to jail
(MH-Prison Population versus MH-General Population) associated with SUDs (no AUD and
no DUD, presence of AUD or DUD, presence of AUD and DUD) and the presence of
psychiatric disorders. A logistic regression model was performed: the outcome was the presence
of at least one psychiatric disorder (mood disorder, anxiety disorder, PTSD, psychotic
syndrome, or high suicide risk), and an interaction term was introduced (SUD * sample). The
model was adjusted for all covariates. Co-occurring psychiatric disorders were also described
for each category (no AUD and no DUD, presence of AUD without DUD, presence of DUD
without AUD, presence of both AUD and DUD), and in each sample.
There was no collinearity in the models: all variance inflation factors (VIFs) were
under 2.5 [27].
Data analysis was performed using R 3.6.1. The significance level was set at α = 0.05,
and all tests were 2-tailed.

12
Results
Sample characteristics
All the socio-demographic characteristics collected differed significantly between the
two groups (see Supplementary Table 1). Overall, individuals in the MH-Prison Population
sample were younger, more precarious, and more often single. They were also more likely to
report a history of migration and trauma exposure.
Rates of past-year susbtance use, and prevalences of AUD and DUD
All rates of past-year substance use and the prevalences of AUD and DUD are provided
in Table 1. Among the 5,793 participants in the MH-General Population group, 5,084 (87.8%)
did not present AUD nor DUD, 205 (3.5%) and 417 (7.2%) presented DUD or AUD,
respectively, and 87 (1.5%) presented both AUD and DUD. Among the 630 participants in the
MH-Prison Population group, 310 (49.2%) did not present AUD nor DUD, 107 (17.0%) and
139 (22.1%) presented AUD or DUD, while 74 (11.7%) presented both AUD and DUD.
Compared to the MH-General Population sample, the rates of past-year use of the
different substances and the prevalences of the different SUDs were significantly higher in the
MH-Prison Population sample, for all substances explored, except for psychedelics (see
Table 1). Alcohol past-year use and AUD were 1.10- and 3.88-fold more frequent in jail,
respectively, whereas past-year illegal drug use and DUDs were 3.48- and 5.74-fold more
frequent, respectively. After adjustment for age category, the prevalences of AUD and DUD
remained 4.15- and 5.20-fold higher in the MH-Prison Population than in the MH-General
Population.

13
Factors correlated with AUD and DUD
Multivariable logistic regression models are presented in Table 2.
Not having a partner, having a history of trauma exposure, and an age between 25 and
44 were associated with a higher risk of AUD. A monthly income level above 1300 €/household
and having a university level of education were associated with a lower risk. There was no
significant association between AUD and employment status or migration history.
Older age was associated with a lower risk of DUD. By contrast, not having a partner,
being unemployed, and reporting a history of migration or trauma exposure were associated
with a higher risk of DUD. No significant association was found regarding income or education
levels.
After adjustment for age, education level, marital status, employment, monthly income,
history of migration, and history of trauma exposure, admission to jail was still associated with
a higher risk of presenting an AUD (aOR=3.80, 95% CI: 2.89-5.01, P<0.001) or a DUD
(aOR=4.25, 95% CI: 3.10-5.84, P<0.001).
Co-occurring psychiatric disorders
Prevalences of co-occurring psychiatric disorders are displayed in Table 3. In the MH-
General Population sample, 24.6% of the participants without AUD and DUD presented at least
one psychiatric disorder, vs 37.7% in the MH-Prison Population sample (P<0.001). This
probability was higher in patients with AUD or DUD, both in the MH-General Population
sample and in the MH-Prison Population sample (respectively 45.3% and 61.1% for AUD,
44.4%, and 58.9% for DUD). This proportion reached 59.8% in the MH-General Population
sample and 91.9% in the MH-Prison Population sample when participants reported both AUD
and a SUD.

14
After adjustment, compared to participants without AUD or DUD, participants with
AUD or DUD (aOR=2.25, 95% CI: 1.89-2.69, P<0.001) or with AUD and DUD (aOR=3.81,
95% CI: 2.44-6.04], P<0.001) were at higher risk of presenting at least one co-occurring
psychiatric disorder. A significant interaction (P=0.007) was found between the SUD status (no
AUD and no DUD, presence of AUD or DUD, presence of AUD and DUD) and the sample
(MH-General Population or MH-Prison Population) on the presence of at least one co-occurring
psychiatric disorder. After stratifying the analysis by sample, in the MH-General Population
group, aORs were 2.21 (95% CI: 1.85-2.64, P<0.001) for AUD or DUD and 3.69 (95% CI:
2.36-5.85, P<0.001) for AUD and DUD, compared to no AUD and no DUD. In the MH-Prison
Population sample, aORs were 2.29 (95% CI: 1.57-3.34, P<0.001) and 16.15 (95% CI: 7.07-
43.85, P<0.001), respectively.
Prevalences of psychiatric disorders among MH-General Population and MH-Prison
Population participants with DUD are described for each type of illegal drug in the
Supplementary Figure 1.

15
Discussion
Our study aimed to assess and compare the rates of past-year substance use, as well as
the prevalence of AUD and DUD, in samples of men from the general population and men
entering jail. We also aimed to compare the associations between AUD and DUD on the one
hand, and co-occurring psychiatric disorders on the other hand.
Our findings reveal that past-year substance use is slightly more frequent in men
entering jail than in the general male population, while crude rates are more than three times
higher for cannabis, more than ten times higher for opioids, more than four times higher for
stimulants, and more than five times higher for other substances. After adjustment for
sociodemographic features, past-year substance use rates remain in the same gap range. In a
previous study in France by Rousselet et al. [28], the reported rates of use before incarceration
were 49.0% of the respondents (vs. 45.6% in our study) for cannabis, 8.9% for heroin (vs. 9.4%
for illegal opioids in our study), and 16.5% for cocaine (vs. 8.6% for stimulants in our study).
The rates of alcohol use could not be compared because different types of use were assessed in
the two studies. Consequently, except for stimulants, our results are in line with previous data,
even if two elements significantly differed between our study and that of Rousselet et al.: the
sample (Rousselet et al. recruited both men and women) and the study design (the interview
was conducted during the third month of incarceration in the work of Rousselet et al.).
Similarly, the crude and adjusted prevalences of AUD and cannabis use disorder were
at least four times higher in the jail sample than in the general population, while the adjusted
prevalence rate of opioid use disorder was 21 times higher, and that of stimulant use disorder 7
times higher. In a previous study performed on French male newcomers in jail, Sarlon et al.
[29] found a crude rate of 16.1 % (95% CI: 11.6-20.6) for alcohol dependence, which may
appear relatively consistent with our rate of 33.8% for AUD, as the diagnostic criteria for
alcohol dependence are more restrictive than those of AUD. The same authors also found a

16
combined prevalence of any drug dependence of 17.6 (95% CI 12.9-22.3), while our estimate
was 28.7%. For the same reason, we believe that the results of both studies are consistent. An
older study by Falissard et al. [30] found much lower prevalences, that is, 8.7% for alcohol
dependence and 8.9% for drug dependence, but this study was conducted during detention and
not at the time of admission in jail, which can explain such differences. Our results are also
consistent with previous international studies in the field. The more recent meta-analysis
exploring AUD and DUD in men entering jail [3] identified a pooled prevalence estimate of
26% (95% CI = 23-30) for AUD (33.8% in our study), and a pooled prevalence estimate of
30% (95% CI = 22-38) for DUD (28.7% in our study).
Our study was the first to specifically assess the associations between AUD and DUD
on the one hand, and co-occurring psychiatric disorders on the other hand. In the non-
incarcerated population, while the occurrence of at least one psychiatric disorder was found in
24.6% of individuals with no AUD nor DUD, it was estimated to reach 44.4% in individuals
with DUD, 45.3% in individuals with AUD, and 59.8% in those with both disorders. In the jail
sample, the occurrence of at least one psychiatric disorder reached 37.7%, 58.9, 61.1%, and
91.9%, respectively. These results highlight how common psychiatric disorders are among
people with SUD, especially among those with both AUD and DUD, regardless of the
imprisonment status [31,32]. They also support a significant difference between the general and
jail populations in each case. Gaps were substantial for anxiety and mood disorders, which were
almost two-fold more frequent in individuals with comorbid AUD and DUD in jail than in those
outside jail.
All these findings are consistent with the dual disorder approach, which postulates that
psychiatric disorders, addictive disorders, and impaired socioeconomic conditions co-occur in
the most vulnerable populations [33]. In this regard, incarceration must be seen as an additional
vulnerability factor that substantially increases the risk of exhibiting a co-occurring psychiatric

17
disorder in individuals with AUD and/or DUD. As a result, the vast majority of individuals with
AUD and/or DUD in jail also display at least one co-occurring psychiatric disorder. A recent
Canadian study showed that the proportion of people with co-occurring mental health needs
and SUDs increased markedly from 15% in 2009 to 32% in 2017 [34]. This suggests that the
treatment of addictive disorders should be associated with a systematic psychiatric examination
and subsequent treatment, especially in jail, but it remains insufficiently done in practice [35].
Several limitations of our study should be acknowledged. First, our study relied only on
self-report, and drug tests were not performed, which can be a source of underreporting.
Moreover, there was a substantial time gap between the MH-Prison Population and MH-
General Population studies, which may have induced some biases, even if it is relatively
unlikely that major changes have occurred between the two periods regarding the rates of
substance use, SUDs, and psychiatric disorders. Second, the judicial history was not explored
in the MH-General Population study interviewees, suggesting that some of the participants
could have been recently released from prison. Similarly, the index offense was not noted in
the MH-Prison Population sample, which did not allow for exploring the relationship between
the reason for incarceration and SUDs. Third, although the statistical model used is appropriate,
ORs should not be considered as relative risks when interpreting the results. Indeed, as the
prevalence of the outcomes is quite high, the ORs overestimate the relative risk. The last
limitation was that the territory explored in the two studies covered only one French region,
located in the extreme north of the country. Consequently, some substantial differences with
more southern French regions, as well as overseas regions, may have been missed, suggesting
that our findings cannot be representative of the entire French general and jail populations.

18
Conclusion
Our study found that men entering jail were much more likely to report past-year
substance use, or to meet a diagnosis of AUD and/or DUD, than men in the general population
of the same French region. These differences persisted after adjustment for sociodemographic
characteristics. Moreover, we found that co-occurring psychiatric disorders were more frequent
in incarcerated subjects with DUD and/or AUD, compared with men recruited in the general
population, suggesting that incarceration constitutes an independent vulnerability factor for
dual disorders. This supports the need for a systematic assessment and treatment of psychiatric
disorders in men entering jail and diagnosed with AUD and/or DUD.

19
Acknowledgement
The authors thank XXXX for her help in editing the manuscript.
Statement of Ethics
The authors assert that all procedures contributing to this work comply with the ethical
standards of the relevant national and institutional committees on human experimentation and
with the Helsinki Declaration of 1975, as revised in 2008.
Legal authorization was obtained by the “XXXX” – authorization number XXXX, the “XXXX”
– authorization number XXXX, and the “XXXX” – authorization number XXXX.
All interviewees provided written informed consent.
Conflict of Interest Statement
The authors declare no conflicts of interest.
Funding Sources
The Mental Health in the Prison Population survey was supported by the XXXX.
Author Contributions
XXXX participated in the conception and design of the study; XXXX participated in the
acquisition of data; XXXX performed the analyses; XXXX wrote the first draft of the
manuscript. XXXX participated in the writing and revision of the successive drafts of the
manuscript and approved the final version.

20
Data Availability Statement
The data on which this manuscript is based are not publicly available. However, data from
Mental Health-Prison Population (MH-Prison Population) are available upon request, under the
supervision of the principal investigator (PI) of the study. Access to data needs to be subjected
to a formal data sharing agreement to ensure that any other researcher accessing the data follows
appropriate ethical standards. The PI of MH-Prison Population (Pr. XXXX) can be contacted
(XXXX) at all times to request data: researchers can submit a research plan, describing its
background, research questions, variables to be used in the analyses, and an outline of the
analyses. If such a request is approved, a written agreement will be signed stating that the data
will only be used for addressing the agreed research questions, and not for other purposes.

21
References
1 Penal Reform International. Global Prison Trends 2020 [Internet]. Penal Reform Int. 2020
[cited 2020 Apr 28]. Available from: https://www.penalreform.org/resource/global-
prison-trends-2020/
2 Fazel S, Hayes AJ, Bartellas K, Clerici M, Trestman R. The mental health of prisoners: a
review of prevalence, adverse outcomes and interventions. Lancet Psychiatry. 2016
Sep;3(9):871–81.
3 Fazel S, Yoon IA, Hayes AJ. Substance use disorders in prisoners: an updated systematic
review and meta-regression analysis in recently incarcerated men and women. Addict
Abingdon Engl. 2017 Oct;112(10):1725–39.
4 Rivlin A, Hawton K, Marzano L, Fazel S. Psychiatric disorders in male prisoners who
made near-lethal suicide attempts: case-control study. Br J Psychiatry J Ment Sci. 2010
Oct;197(4):313–9.
5 Radeloff D, Hövel M ten, Brennecke G, Stoeber FS, Lempp T, Kettner M, et al. Suicide
after reception into prison: A case-control study examining differences in early and late
events. PLOS ONE. 2021 Aug;16(8):e0255284.
6 Kinner SA, Forsyth S, Williams G. Systematic review of record linkage studies of
mortality in ex-prisoners: why (good) methods matter. Addict Abingdon Engl. 2013
Jan;108(1):38–49.
7 Chang Z, Lichtenstein P, Larsson H, Fazel S. Substance use disorders, psychiatric
disorders, and mortality after release from prison: a nationwide longitudinal cohort study.
Lancet Psychiatry. 2015 May;2(5):422–30.
8 Chang Z, Larsson H, Lichtenstein P, Fazel S. Psychiatric disorders and violent
reoffending: a national cohort study of convicted prisoners in Sweden. Lancet Psychiatry.
2015 Oct;2(10):891–900.
9 Steiner B, Wooldredge J. Inmate Versus Environmental Effects On Prison Rule
Violations. Crim Justice Behav. 2008 Apr;35(4):438–56.
10 Oser CB, Knudsen HK, Staton-Tindall M, Taxman F, Leukefeld C. Organizational-level
correlates of the provision of detoxification services and medication-based treatments for
substance abuse in correctional institutions. Drug Alcohol Depend. 2009 Aug;103 Suppl
1:S73-81.
11 Wakeman SE, Rich JD. Pharmacotherapy for substance use disorders within correctional
facilities. Oxford Textbook of Correctional Psychiatry. Oxford University Press; 2015.
12 Mundt AP, Baranyi G, Fazel S. Incomplete discussion of bipolar disorder and comorbid
substance use disorder - Authors’ reply. Lancet Glob Health. 2019;7(7):e847.
13 Mir J, Kastner S, Priebe S, Konrad N, Ströhle A, Mundt AP. Treating substance abuse is
not enough: comorbidities in consecutively admitted female prisoners. Addict Behav.
2015 Jul;46:25–30.

22
14 Abram KM, Zwecker NA, Welty LJ, Hershfield JA, Dulcan MK, Teplin LA. Comorbidity
and continuity of psychiatric disorders in youth after detention: a prospective longitudinal
study. JAMA Psychiatry. 2015 Jan;72(1):84–93.
15 Ignatyev Y, Baggio S, Mundt AP. The Underlying Structure of Comorbid Mental Health
and Substance Use Disorders in Prison Populations. Psychopathology. 2019;52(1):2–9.
16 Fovet T, Plancke L, Amariei A, Benradia I, Carton F, Sy A, et al. Mental disorders on
admission to jail: A study of prevalence and a comparison with a community sample in
the north of France. Eur Psychiatry J Assoc Eur Psychiatr. 2020 Apr;63(1):e43.
17 Bartlett A, Hollins S. Challenges and mental health needs of women in prison. Br J
Psychiatry J Ment Sci. 2018;212(3):134–6.
18 Bellamy V, Roelandt J-L, Caria A. Premiers résultats de l’enquête Santé mentale en
population générale : images et réalités. Inf Psychiatr. 2005 Apr;81(4):295–304.
19 Caria A, Roelandt J-L, Bellamy V, Vandeborre A. ["Mental Health in the General
Population: images and realities (MHGP)": methodology of the study]. L’Encephale.
2010;36(3 Suppl):1–6.
20 Amad A, Guardia D, Salleron J, Thomas P, Roelandt J-L, Vaiva G. Increased prevalence
of psychotic disorders among third-generation migrants: results from the French Mental
Health in General Population survey. Schizophr Res. 2013 Jun;147(1):193–5.
21 Pignon B, Geoffroy PA, Thomas P, Roelandt J-L, Rolland B, Morgan C, et al. Prevalence
and clinical severity of mood disorders among first-, second- and third-generation
migrants. J Affect Disord. 2017 Mar;210:174–80.
22 Pignon B, Amad A, Pelissolo A, Fovet T, Thomas P, Vaiva G, et al. Increased prevalence
of anxiety disorders in third-generation migrants in comparison to natives and to first-
generation migrants. J Psychiatr Res. 2018 Jul;102:38–43.
23 Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, et al. The Mini-
International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a
structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry.
1998;59 Suppl 20:22-33;quiz 34-57.
24 Lecrubier Y, Sheehan DV, Weiller E, Amorim P, Bonora I, Sheehan KH, et al. The Mini
International Neuropsychiatric Interview (MINI). A short diagnostic structured interview:
reliability and validity according to the CIDI. Eur Psychiatry. 1997 ed;12(5):224–31.
25 Sheehan D, Lecrubier Y, Harnett Sheehan K, Janavs J, Weiller E, Keskiner A, et al. The
validity of the Mini International Neuropsychiatric Interview (MINI) according to the
SCID-P and its reliability. Eur Psychiatry. 1997 Jan;12(5):232–41.
26 Porta M. A Dictionary of Epidemiology. Oxford University Press; 2016
27 Johnston R, Jones K, Manley D. Confounding and collinearity in regression analysis: a
cautionary tale and an alternative procedure, illustrated by studies of British voting
behaviour. Qual Quant. 2018;52(4):1957–76.

23
28 Rousselet M, Guerlais M, Caillet P, Geay BL, Mauillon D, Serre P, et al. Consumption of
psychoactive substances in prison: Between initiation and improvement, what trajectories
occur after incarceration? COSMOS study data. PLOS ONE. 2019 Dec;14(12):e0225189.
29 Sarlon E, Duburcq A, Neveu X, Morvan-Duru E, Tremblay R, Rouillon F, et al.
Imprisonment, alcohol dependence and risk of delusional disorder: a cross-sectional
study. Rev Epidemiol Sante Publique. 2012 Jun;60(3):197–203.
30 Falissard B, Loze J-Y, Gasquet I, Duburc A, de Beaurepaire C, Fagnani F, et al.
Prevalence of mental disorders in French prisons for men. BMC Psychiatry. 2006
Aug;6:33.
31 Castillo-Carniglia A, Keyes KM, Hasin DS, Cerdá M. Psychiatric comorbidities in
alcohol use disorder. Lancet Psychiatry. 2019 Dec;6(12):1068–80.
32 Comorbidity of substance use and mental health disorders in Europe (Perspectives on
drugs) | www.emcdda.europa.eu [Internet]. [cited 2022 Jun 13]. Available from:
https://www.emcdda.europa.eu/publications/pods/comorbidity-substance-use-mental-
health_en
33 Szerman N, Peris L. Precision Psychiatry and Dual Disorders. J Dual Diagn. 2018
Dec;14(4):237–46.
34 Butler A, Nicholls T, Samji H, Fabian S, Lavergne MR. Prevalence of Mental Health
Needs, Substance Use, and Co-occurring Disorders Among People Admitted to Prison.
Psychiatr Serv. 2021 Nov;appi.ps.202000927.
35 Reingle Gonzalez JM, Connell NM. Mental Health of Prisoners: Identifying Barriers to
Mental Health Treatment and Medication Continuity. Am J Public Health. 2014
Dec;104(12):2328–33.

24
Table 1. Rates of past-year substance use and prevalences of substance use disorders in the MH-General Population and the MH-General Population samples
Crude rates in MH-General
Population
(N = 5,793)
Crude rates in MH-Prison
Population
(N = 630)
p
Age-standardized rates in
MH-Prison Population
[95% CI]
N
% [95% CI]
N
% [95% CI]
Past-year substance uses
Alcohol
3,266
56.4 [55.1-57.7]
391
62.1 [58.1-65.8]
0.007
63.3 [59.4-67.1]
Illegal drugs†
At least one type of illegal drug
833
14.4 [13.5-15.3]
315
50.0 [46.1-53.9]
<0.001
42.1 [38.2-46.1]
Cannabis
803
13.9 [13.0-13.9]
287
45.6 [41.6-49.5]
<0.001
38.0 [34.3-42.0]
Opioids
43
0.7 [0.5-1.0]
59
9.4 [7.2-12.0]
<0.001
10.9 [8.6-13.7]
Stimulants
112
1.9 [1.6-2.3]
54
8.6 [6.5-11.1]
<0.001
8.7 [6.7-11.3]
Psychedelics
29
0.5 [0.3-0.7
7
1.1 [0.5-2.4]
0.095
0.8 [0.3-1.9]
Other
26
0.4 [0.3-0.7]
16
2.5 [1.5-4.2]
<0.001
2.9 [1.8-4.6]
Substance use disorders
Alcohol
504
8.7 [8.0-9.5]
213
33.8 [30.1-37.7]
<0.001
36.1 [32.4-40.0]
Illegal drugs
At least one type of illegal drug
292
5.0 [4.5-5.6]
181
28.7 [25.2-32.5]
<0.001
26.0 [22.7-29.7]
Cannabis
274
4.7 [4.2-5.3]
151
24.0 [20.7-27.5]
<0.001
21.0 [17.9-24.4]
Opioids
21
0.4 [0.2-0.6]
43
6.8 [5.0-9.1]
<0.001
8.4 [6.4-10.9]
Stimulants
46
0.8 [0.6-1.1]
33
5.2 [3.7-7.3]
<0.001
5.6 [4.0-7.8]
Psychedelics
10
0.2 [0.1-0.3]
4
0.6 [0.2-1.7]
0.056
0.4 [0.1-1.5]
Other
11
0.2 [0.1-0.3]
13
2.1 [1.1-3.6]
<0.001
2.6 [1.5-4.2]
† The type of substance was not specified for 17 participants with past-year illegal drug use (but not DUD) in the MH-General Population sample. No other data was missing.

25
Table 2. Results of the multivariate logistic regression models assessing the factors correlated with alcohol or
illegal drug use disorders (AUD or DUD)
AUD†
DUD‡
aOR$
[95% CI]
p
aOR$
[95% CI]
p
Group
MHGP
Ref
Ref
MHPP
3.80
[2.89-5.01]
<0.001
4.25
[3.10-5.84]
<0.001
Age (years)
<25
Ref
Ref
25-34
1.43
[1.11-1.86]
0.006
0.60
[0.46-0.77]
<0.001
35-44
1.55
[1.17-2.07]
0.002
0.39
[0.28-0.54]
<0.001
>44
1.07
[0.81-1.43]
0.633
0.08
[0.05-0.13]
<0.001
Education level
No education / Primary level
Ref
Ref
Secondary level
0.83
[0.68-1.02]
0.076
0.89
[0.70-1.13]
0.345
University level
0.58
[0.44-0.75]
<0.001
0.79
[0.58-1.06]
0.128
Marital status
Married or co-residing with partner
Ref
Ref
Single
1.41
[1.15-1.74]
0.001
1.61
[1.25-2.08]
<0.001
Separated / divorced / widowed
1.60
[1.25-2.04]
<0.001
1.56
[1.03-2.30]
0.030
Employment
Yes
Ref
Ref
No
0.97
[0.80-1.17]
0.754
1.44
[1.14-1.82]
0.002
Monthly income level (€/household)
<534
Ref
Ref
534-840
0.87
[0.64-1.19]
0.405
1.02
[0.71-1.46]
0.906
840-1300
0.84
[0.61-1.14]
0.264
0.94
[0.64-1.37]
0.741
>1300
0.66
[0.48-0.91]
0.010
0.79
[0.56-1.13]
0.191
History of trauma exposure
No
Ref
Ref
Yes
1.81
[1.53-2.14]
<0.001
2.15
[1.74-2.65]
<0.001
History of migration
No
Ref
Ref
Yes
1.01
[0.84-1.21]
0.939
1.38
[1.12-1.71]
0.003
†Reference group: no AUD
‡Reference group: no DUD
$Adjusted for all the variables listed in the tabl

26
Table 3. Prevalence of psychiatric disorders according to the presence of alcohol use disorder (AUD) or illegal drug use disorder (DUD), in each sample (MH-General
Population and MH-Prison Population)
No AUD or DUD
AUD without DUD
DUD without AUD
AUD and DUD
MH-General
Population
(n=5,084)
MH-Prison
Population
(n=310)
p
MH-General
Population
(n=417)
MHPP
(n=139)
p
MH-General
Population
(n=205)
MHPP
(n=107)
p
MH-General
Population
(n=87)
MHPP
(n=74)
p
At least one
24.6
[23.4-25.8]
37.7
[32.3-43.4]
<0.001
45.3
[40.5-50.2]
61.1
[52.4-69.1]
0.002
44.4
[37.5-51.5]
58.9
[49.0-68.2]
0.021
59.8
[48.7-70.0]
91.9
[82.6-96.7]
<0.001
PTSD
0.5
[0.3-0.7]
2.2
[1.0-4.7]
0.002
1.9
[0.9-3.9]
4.3
[1.8-9.5]
0.126
1.5
[0.4-4.6]
4.7
[1.7-11.1]
0.128
1.1 [0.1-7.0]
16.2
[9.0-27.0]
0.001
High
suicidal risk
1.2
[0.9-1.5]
5.8
[3.6-9.2]
<0.001
5.3
[3.4-8.0]
15.8
[10.4-23.2]
<0.001
3.9
[1.8-7.8]
12.1
[6.8-20.2]
0.012
19.5
[12.1-29.7]
29.7
[20.0-41.6]
0.187
Psychotic
syndromes
2.0
[1.6-2.4]
3.2
[1.6-6.0]
0.219
6.9
[4.7-9.9]
7.9
[4.2-14.0]
0.850
3.9
[1.8-7.8]
7.5
[3.5-14.7]
0.276
10.3
[5.1-19.1]
20.3
[12.2-31.6]
0.123
Anxiety
disorders
17.8
[16.8-18.9]
30.0
[25.0-35.5]
<0.001
30.7
[26.3-35.4]
48.2
[39.7-56.8]
<0.001
33.6
[27.3-40.6]
49.5
[39.8-59.3]
0.009
44.8
[34.2-55.8]
78.4
[67.0-86.8]
<0.001
Mood
disorders
11.1
[10.2-12.0]
20.6
[16.3-25.6]
<0.001
29.2
[24.9-33.9]
38.1
[30.1-46.7]
0.065
25.4
[19.7-32.0]
29.9
[21.6-39.6]
0.469
37.9
[27.9-49.0]
63.5
[51.4-74.1]
0.002