Factors associated with risk related to the use of psychoactive substances by men deprived of their liberty

Abstract Objective: to evaluate the factors associated with risk related to the use of psychoactive substances in male inmates of a prison in a city in the South of Brazil. Method: a cross-sectional data from 220 men deprived of liberty, inmates of a provisional custody institution in the State of Paraná, collected with a screening instrument and questionnaire. Binary logistic regression and odds ratio analysis were used to verify associations between risk related to substance use and socio-demographic characteristics of living conditions before incarceration and current incarceration. Results: the adjusted model revealed association of consumption with skin color brown/black and yellow, those who had only one parent responsible until age 15, age at first arrest 18 or older, professing religion, working before arrest, owning their own house, living alone, receiving visitors in prison. Conclusion: the identified factors are useful to insert effective treatment proposals and reduce the gaps and social vulnerability existing in prison.


Introduction
The prison environment has occupied a prominent place in public policies worldwide due to the rapid pace of growth of the prison population. In the world, there are more than ten million people living in prison, and Brazil ranks third among the countries with the largest prison populations in the world, with about 730,000 persons deprived of liberty (PDLs) (1)(2) .
Investigating the health conditions of PDLs represents a challenge to researchers, since this population is considered by public opinion as undeserving of any assistance, and the problems identified are seen as punishment for previous acts. However, it is understood that a hostile and unhealthy prison environment can hinder the subsequent re-socialization process (3) .
The prison environment is considered detrimental to the physical and emotional health conditions of PDLs, culminating not only in the deprivation of liberty, but also of dignity (4) . Prisons have overcrowded facilities, increasing the risk of exposure to various untreated or undetected pathologies, violence, and psychoactive substance abuse (PASs) (5) .
Lifelong use of illicit substances and the consequent chemical dependency are a reality for more than 50% of PDLs (6)(7) . Many PDLs report being under the influence of PAS at the time they committed the crime for which they were arrested or that the reason for arrest was related to offenses related to trafficking, possession or consumption of PASs (8) .
As a consequence, the worldwide rate of chemical dependency-related mental disorders from PASs has increased significantly in recent decades, reaching approximately 29.5 million people in 2015 (9) . In this context, it is observed that PASs are often present in prison systems around the world, being particularly used at the beginning, as a way for PDLs to deal with the evils of incarceration, such as overcrowding, unhealthy conditions, exposure to violence, lack of health care and the breakdown of family ties (10)(11)(12) . With continued and prolonged use, PASs cease to be a means to survival and the consequent development of addiction becomes an end in itself (13) .
It is a consensus in the national and international literature that the addicts of PAS are overrepresented in the prison populations and with similar characteristics: low socioeconomic status, low education and with physical and mental health problems (12,(14)(15)(16) . Because incarceration represents a constant challenge to be faced by PDLs, the public health field must be concerned with prisons as a cause of health inequities (16) .
Economic, family, housing, skin color, age, among others, are social determinants of health that directly impact the PDLs (17)(18) . Therefore, the relationships between incarceration, PASs, and social determinants of health are urgently needed evidence to improve quality of life and subsequent re-socialization for PDLs (17) . Although substance abuse in PDLs is estimated to be ten times more prevalent than in the general population, problems with these substances are not always detected in prisons (12) . In addition, the perception of belonging to  (19) .
It is known that the consumption of PASs is allowed in the prison environment in the case of prescribed psychotropic drugs and tobacco, but any other PASs are prohibited during incarceration (4) . However, data on this consumption in PDLs is still quite scarce. It is believed that the lack of information on its circulation and consumption in prisons may be related to the complexity of discussing these phenomena in public security institutions (10) .
Because it is a veiled and denied situation, it is believed that PDL do not receive mental health care suited to the premises of harm reduction because the data on the number of PAS users in prisons and the types of PASs used are most likely under-reported (13) .
Given this context, it is critical to identify predictors and risk factors for PAS use by PDLs in order to correct the current paucity of literature and inform appropriate prevention and harm minimization responses. Knowledge of predictors and risk factors for PAS use by PDLs may allow for the prediction of PAS use consequent to the generation of data used to inform specific policy and prevention options for high consumption in PDLs.
Therefore, the objective of this study was to evaluate the consumption of PAS and associated factors among male prisoners in a prison in a city in the South of Brazil.

Study design
This is a cross-sectional study, carried out with men deprived of liberty, inmates of a provisional maximum security prison unit in a city in northwestern Paraná. The guidelines for strengthening the reporting of observational studies in epidemiology (STROBE) were followed (20) .

Setting where the data collection took place
The research setting was a temporary custody house in a medium-sized municipality located in the northwest of the State of Paraná, Brazil. The institution was opened in 2008, being a maximum security penal establishment whose purpose is to allocate vacancies only to provisional PDLs awaiting criminal conviction, specifically for the male population. However, due to the shortage of vacancies in the state penitentiary of reference, due to overcrowding, the penal unit absorbs provisional PDLs and also those already convicted.

Period
Data was collected in the months of June to November 2019, in the morning and afternoon periods.

Population
The study selected men deprived of their freedom who were inmates of a provisional custody institution. In the month before data collection started, the unit housed 1183 inmates; 535 were convicted and 648 were temporary.

Selection criteria
As the house of custody is intended for PDLs without criminal conviction, we considered as a selection criterion only men in provisional regime (prison management software SPR, v2).
Those with clinical diagnoses related to Neurology and Psychiatry and/or with cognitive limitations that hindered communication and responses to the interviews were excluded (4) and with a prison time of less than 25 days (1).

Sample definition
With the list made available by the institution containing all men deprived of liberty and considering that the population is finite (643 PDLs), stratified sampling was carried out, with an estimation error of 5%, Confidence Interval of 95% and prevalence of 30% (21)(22)(23) , resulting in a minimum sample size of 216 people.
After the sample calculation, a random and stratified drawing was made of 160 cells that house, on average, eight people. This way, all the PDLs considered as provisional had chances to belong to the sample and, at the end of the selection process, the final sample of the study consisted of 220 people.

Study variables
To assess the risk related to the use (RRU) of PASs in PLWH, information was collected regarding three groups of independent variables. The first group contained the sociodemographic characteristics: age in years and categorized (19 to 39 years and 40 to 64 years); selfreported skin color, which was classified as white and other (brown/black or yellow); having a partner (no and yes); religious beliefs (yes and no), and children (yes and no). In the second group, living conditions before imprisonment were considered: housing condition, classified as owned, rented, and others (borrowed, relatives, or homeless); lived alone (yes and no); worked before imprisonment (yes and no); family income (no income and with income); family member responsible until age 15, categorized as both parents, only one parent, and others (other relatives or none), and age at

Data collection instrument
A structured interview script, on paper and pen, was administered to all participants, taking approximately 40 minutes to complete. It was composed of two modules: and other PASs (24)(25) .
The instrument was adapted to the Brazilian culture and is composed of eight questions, easy to apply, which address the frequency of PAS use in life and in the last three months, problems related to use, concern about the use by people close to the user, impairment in the execution of expected tasks, unsuccessful attempts to stop or reduce use, feeling of compulsion, and injecting use. Each response corresponds to a score, the sum total of which can range from zero to 39 is indicative of abuse, and brief intervention and/or counseling is recommended. High risk (score of 27 or higher) is suggestive of dependence, with referral to intensive treatment (24)(25)(26) . After applying the instrument, individuals who have never used any of the PASs are considered to be at no risk.

Data collection
Data collection was carried out by a single researcher, a nurse from the Postgraduate Program in

Data treatment and analysis
After collecting the information, the data were compiled in electronic spreadsheets. Next, descriptive analysis (mean, standard deviation, median, and absolute and relative frequencies) was performed for the variables that characterized the PDLs (sociodemographic, living conditions before incarceration and current incarceration).
The presence (low, moderate, and high) or absence (none) of the RRU of PASs in PDLs after ASSIST screening was considered as the outcome variable. Univariate and multiple binary logistic regression models were employed to determine the factors associated with the presence of RRU. The stepwise both method was used for the selection of variables and fitting of the final models. The adequacy of these models was verified with the analysis of quantile randomized residuals (27) and collinearity was tested with the variance inflation factor (VIF). Associations were estimated by calculating the odds ratio (OR), adopting the 95% Confidence Interval (CI) as a measure of accuracy (28) . The analyses were performed in R software, version 4.0.4 (29) .

Ethical aspects
After the appreciation of the Permanent Committee

Results
The mean age (years) of males was 30.9, with a standard deviation of 10.1 and a median of 29, with a minimum age of 19 and a maximum of 64. There was a predominance of adult subjects, aged 19 to 59 years (n=216), when compared to the elderly, aged 60 to 64 years (n=4). The sociodemographic profile and living conditions before incarceration and current incarceration of the 220 men are detailed in Table 1

Discussion
The main findings of this study were: the  The imprisonment experience is a complex process and can cause a high prevalence of mental disorders, endangering the health of those who are incarcerated, perpetrating self-destructive behaviors (4,30) .
The consumption of PAS may be related to the very socialization in the prison environment, facilitating the insertion of inmates into dominant social groups in prison.
In addition, the use of PAS may act as a defense and escape mechanism for the mental health of detainees with presumable psychopathological worsening, since the addiction is maintained even after the completion of the sentence in the resocialization process (4,13) .
Given this scenario, we identified a significant prevalence of substance abuse consumption in the lives of PDLs, mainly alcohol, tobacco, marijuana and cocaine/ crack. Marijuana is the most commonly consumed illicit substance, corroborating the national and international literature (14,31) , with prevalence rates much higher than those observed among the general Brazilian population (marijuana 7.7%, cocaine 3.1%, crack 0.9%) (32) .
Marijuana was the most commonly used illicit drug, followed by cocaine/crack and inhalants, and about a quarter had used hypnotics, hallucinogens, opioids and amphetamines or ecstasy, corroborating international studies that point to marijuana as the most commonly used illicit drug among PDLs (33) .
Regarding RRU of SPAs of abuse, moderate marijuana use was identified in this study, with risk related to low age at first arrest. One study found similar results whose marijuana use was reported by 67.5% of PDLs with onset at the age of ten to 15 years (34) . Cannabis is the most commonly consumed illicit substance and can act as a "gateway" to other, heavier drugs (14) .
Corroborating the findings of this research, a study conducted in France concluded that substance abuse in the prison environment may be related to the high concentration of arrests for PAS-related crimes, low socioeconomic status, and frequent psychiatric disorders in PDLs (14) . Low-and middle-income countries, such as Brazil, may have a prevalence of PAS abuse and dependence of 25% among PDLs (35) . Research conducted in Ethiopian prisons identified that lack of social support, living in urban areas, psychopathy, and family history of substance use are associated with abuse PAS use in PDLs (15) . Tobacco has long been considered part of the prison culture, and the smoking situation among PDLs is more serious (37) . In this study, a moderate risk for tobacco use was identified related to age at first arrest, skin color, and family member responsible for care until age 15.
A North American study demonstrated that adolescents raised by both parents is a protective factor against the use of tobacco, alcohol, and illicit PASs (11) .
These findings corroborate research that identified increased frequency of tobacco use among prisoners on the grounds of coping with the stress associated with incarceration (38) . The increased consumption of PAS by prisoners in Ecuador was also associated with incarceration (39) , portraying the need to address this issue in the prison environment in order to plan efficient and effective actions with PDLs.
Regarding alcohol, its consumption in the prison environment showed low risk and was associated with the practice of religiosity. The role that religion and spirituality play in the cessation of criminal behavior and the use of PASs is not yet fully understood, but suggests a relatively high importance in substance use in the prison environment, particularly in relation to alcohol and cocaine (40) . Another Brazilian study also found that inhaled cocaine, at moderate and high levels, had a statistically significant association with the variables not professing religion, risky sexual behavior, age 18 to 34 years, and living with a drug user (41) .
The family context emerges as preponderant in the discussion of substance abuse and the family emerges as the first circle of socialization, internalization of emotions and behaviors that will be experienced in other environments. A study conducted in Greek prisons also found associations between sociodemographic variables with the consumption of PAS as the beginning of consumption at early ages, low education and performance of work activities (34) .
The use of injectable PASs leads the individual to an increased risk of contracting infectious diseases, such as hepatitis C and human immunodeficiency virus. In this study, the use of injectable PASs was lower when compared to other international studies (14,42) . However, an increased prevalence of drug use during incarceration was observed. Approximately 15% of PDLs used medications in prison, showing that the main trend was an increase in the consumption of controlled drugs and a decrease in the consumption of other illicit substances of abuse, used as justifications to forget the condition of incarceration (14,42) . However, the fact that there is still abuse of PASs in prison settings still portrays a reality far from ideal.
The limitation of this study is its cross-sectional design, and it is not possible to establish temporality or causality. Another limitation is due to the fact that specific variables related to mental disorders were not included, addressing only those related to chemical dependence.

Conclusion
In this study, PDLs showed high prevalence of PASs use in life, and the risks related to use were moderate for tobacco and marijuana in the prison environment.