key: cord-0796414-o9qeqv3i authors: Tindimwebwa, Linda; Ajayi, Anthony Idowu; Adeniyi, Oladele Vincent title: Prevalence and Demographic Correlates of Substance Use among Adults with Mental Illness in Eastern Cape, South Africa: A Cross-Sectional Study date: 2021-05-19 journal: Int J Environ Res Public Health DOI: 10.3390/ijerph18105428 sha: 22b9e462e2e5327b5a377292682029294e449058 doc_id: 796414 cord_uid: o9qeqv3i This study reports on the prevalence and demographic correlates of substance use among individuals with mental illness in the Eastern Cape, South Africa. This cross-sectional study was conducted in the Outpatient Clinic of a large hospital in the Eastern Cape, South Africa. A pre-validated tool on alcohol and psychoactive drug use was administered to 390 individuals with mental illness. Multivariable logistic regression models were fitted to explore the demographic correlates of alcohol and psychoactive drug use. Of the total participants (N = 390), 64.4% and 33.3% reported lifetime (ever used) and past-year use of alcohol, respectively, but the prevalence of risky alcohol use was 18.5%. After adjusting for relevant covariates, only male sex, younger age, and rural residence remained significantly associated with risky alcohol use. The prevalence of ever-use and past-year use of psychoactive substances was 39.7% and 17.4%, respectively. The most common substance ever used was cannabis (37.4%). Male sex, younger age, owning a business, and being unemployed were significantly associated with higher odds of lifetime and past-year use of psychoactive substances. Findings highlight the need for dedicated infrastructure and staff training in the management of these dual diagnoses in the region. Substance use is an important risk factor for morbidity, disability, and premature death globally [1] . Besides the adverse consequences on health, alcohol and psychoactive drugs also create a significant burden on productivity and economic and social aspects of individuals, as well as their families and communities [2] . It was estimated that in 2017, 271 million people (5%) of the global population between 15 and 64 years of age had used psychoactive drugs in the previous twelve months [3] . A national survey of South African adults showed a prevalence of 33.1% for past-year alcohol use, which was higher than the global average of 32.5% [4, 5] . In addition, a total of 62,300 alcohol-related deaths were reported in 2015 [6] . Several studies have reported a higher prevalence (range of 21.3-55.6%) of substance use disorders in individuals with mental illness compared to three to four percent reported in the general population [7, 8] . In Nigeria, two studies on the use of psychoactive substances in mentally ill individuals showed prevalences between 21.3% and 29.3% [9, 10] . In Finland, Karpov et al. (2017) [11] reported a prevalence of 27.5%. The Epidemiologic Catchment Area Study, a large study carried out in the United States, found a prevalence rate of co-morbidity of 29% [12] . A study done in the Western Cape of South Africa showed a prevalence rate of 55.6% [13] . All of the studies had a confidence interval of 95%. Substance use and mental illness are strongly inter-related [14, 15] . Several theories have been advanced to explain the bi-directional and complex relationship between these dual diagnoses, which often co-exist in the same individual. The self-medication hypothesis postulates that mental illness might contribute to substance use and addiction due to individuals using specific substances to ameliorate specific symptoms [16] [17] [18] [19] . Another theory suggests that individuals with harmful substance use have a higher risk of developing a psychiatric disorder [16, 18, 20] . For psychiatric disorders to evolve in individuals with a history of substance use, the authors of [21] argued that genetic predisposition or environmental factors have crucial roles to play. It has also been hypothesized that substance abuse and psychiatric disorders originate from a common, overlapping genetic vulnerability in the affected individuals [16, 18] . Bi-directional models suggest that either disorder increases the vulnerability to the other in the form of a feedback loop [18, 20] . The complex relationship between substance use and mental illness negatively impacts the natural course, clinical outcomes, and prognosis of either condition [22, 23] . Substance use in individuals with mental illness leads to poor adherence to treatment, deterioration of condition, recurrent relapses, and hospitalizations [22, 23] . Individuals with co-morbidity of substance and psychiatric disorders are more likely to be unemployed, homeless, and violent and have family and interpersonal relationship instability [22, 23] . They tend to experience victimization, social exclusion, legal problems, and a greater prevalence of acquiring communicable diseases like HIV or Hepatitis, due to their risk-taking behaviors [24] [25] [26] . Cannabis is the most commonly used psychoactive drug globally. In 2017, an estimated 188 million (3.8%) of the global population between 15 and 64 years of age had used cannabis in the previous year [3] . The term opioids include both opiates and non-medical use of prescription opioids. Examples of these are heroin, morphine, codeine, tramadol, and fentanyl. Past year opioid use in the same age group globally in 2017 was estimated at 53.4 million (1.1%), an increase of 56% from 2016. Amphetamines, methamphetamine, in particular, had an estimated 28.9 million (0.6%) past year users globally [3] . A national survey in South Africa looking at the past 3-month use of psychoactive drugs found that cannabis (dagga, weed, zol) was the most used drug with a prevalence of 4.0%, which is higher than the global average. Following cannabis were sedative-hypnotics, examples of which are diazepam (valium), methaqualone (mandrax) and Rohypnol (0.4%). Amphetamines (tik, speed), cocaine (coke, crack, rock), and opiates each had a prevalence of use of 0.3% [8] . Authors [27] Previous studies on alcohol use and harmful drinking in the Buffalo City Municipality and Amathole district in South Africa have focused on the general population [28, 29] , neglecting individuals with mental illness. This study reports on the prevalence and correlates of substance use (risky use of alcohol and psychoactive drugs) among individuals with mental illness attending the Outpatient Clinic of the Cecilia Makiwane Hospital Mental Health Unit (CMHMHU) in East London, Eastern Cape, South Africa. The study's findings could guide the crafting of context-specific interventions and policies on the prevention and continued support for individuals with dual diagnoses of mental illness and substance use. This cross-sectional study was conducted in the Outpatient Clinic of the Cecilia Makiwane Hospital Mental Health Unit (CMHMHU) situated in Mdantsane, East London, between March and June 2020. This department provides both inpatient and outpatient mental health care to the 1,674,637 people living in Buffalo City Metropolitan Municipality (BCMM) and Amathole district in the Eastern Cape Province of South Africa [30] . The department is an academic unit affiliated with Walter Sisulu University, South Africa, for the training of undergraduate and postgraduate students in psychiatry. The inpatient facility consists of 50 beds, 25 male and 25 female. The outpatient unit sees between 13,000 and 15,000 patients a year. The sample size for this study was estimated as 390, using Cochran's formula for cross-sectional study [31] : where z1 − ∝/2 is the standard normal variate at 1.96. p = expected proportion of patients with co-morbidity of substance use and mental illness, based on a previous study [32] , and observation of patients admitted in the wards at CMHMHU, at 50%, which is equal to 0.5. D is the absolute precision and is taken as 0.05. All individuals attending CMHMHU outpatients' clinic were considered eligible for the study. Participants were included if they were 18 years and older, had been on treatment for mental illness for at least a year, and had no violent or aggressive behavior. However, some participants were excluded if they displayed behavior that posed a danger to themselves, others, and property during the study. Such individuals were immediately offered emergency care per departmental protocol. A questionnaire was designed using components of the validated National Insitute on Drug Abuse (NIDA), Rockville, MD, United States Quick Screen [33] and WHO STEPwise questionnaires [34] . The study instrument contained three sections: demographic data, use of alcohol and psychoactive drugs, and the current diagnosis of mental illness. The tool was piloted with ten patients at the study site in order to ascertain its acceptability and feasibility. Using feedback from the participants and the attending clinicians, final adjustments were made to the questionnaire. Participants of the pilot were not included in the main study. Information about the study, its purpose, and the questionnaire were provided to participants verbally and through an information sheet. Doctors were given information sheets regarding the eligibility criteria and trained on how to administer the questionnaire, which was completed after consultation with the participants. To supplement the information obtained from self-report by the participants, medical records were reviewed in order to validate the responses given. Where discrepancies were found, the participants' clinical notes were taken as the true reflection of the use of the substances under inquiry. Additionally, collateral information was used to validate the self-report of the main outcome measures. Participants were recruited consecutively by the attending doctors. Data collection was done over a period of four months, from March to June 2020. Given that this study was conducted during the COVID-19 pandemic, we strictly followed the necessary precautions based on local protocols. Alcohol use was measured in terms of a "standard drink" in order to classify use as risky or non-risky. There is no global consensus for what constitutes a "standard drink", with a study finding significant variations between countries [35] . The World Health Organisation (WHO), Geneva, Switzerland, adopted ten grams of pure alcohol per drink as the definition of a standard drink. In South Africa, one unit of alcohol is twelve grams of pure alcohol per drink. Low-risk drinking is classified as twenty-four grams (2 units) daily for both men and women [35] . Using the definition by the National Institute on Alcohol Abuse and Alcoholism, 2020, risky alcohol use in this study was defined as consumption of four or more standard drinks by men and three or more standard drinks by women daily, or, on a weekly basis, 21 drinks for men and 14 drinks for women. Alcohol use was self-reported and categorized as current use if participants had consumed alcohol in the past twelve months (or past year) [36] and ever-use (lifetime). Psychoactive drugs included in this study were those that could result in dependence and whose use was deemed risky. The selection of substances was informed by the commonly used psychoactive drugs in South Africa reported by [37] . These include dried herbal cannabis (dagga), methaqualone (mandrax), crystal methamphetamine (tik), opioids, and cocaine [37] . In order to accommodate outliers, the category of "others" was included. Use of psychoactive substances was elicited by self-report and categorized as past year use and ever-use (lifetime). Pertinent demographic characteristics considered in the study, age, sex, highest level of education, race, marital status, employment, monthly household income, and area of residence (rural vs. urban) were obtained by self-reporting. The participants' ages (in years) were grouped as 18-25, 26-35, 36-45, 46-55, 56-65, and 66 years or older. Their highest level of education was grouped according to grades, with any post-high school education categorized as tertiary. Household monthly income in rands was categorized as none, R150-2000, R2001-5000, and R5001 or more. Employment status was categorized as none, having a job or business, and social assistance (child support grant, older person's grant, disability grant, grant-in-aid, care dependency grant, war veteran's grant, foster child grant and social distress relief grant). Participants were considered unemployed if they had no work in both the formal or informal sectors. Ethical approval for the study was received from the Walter Sisulu University Research Ethics Committee (reference number 093/2019). Also, the Eastern Cape Department of Health and the clinical governance of Cecilia Makiwane Hospital, Mdatsane, South Africa, granted permission for the researcher to conduct the study. Participants and/or their legal guardians, in the event that the patient was unable to give consent due to diminished capacity, received an information sheet in their preferred language (English or Xhosa) detailing the study's purpose and process. Participants (or guardians) who were willing to participate in the study gave informed consent in writing. We respected participants' rights to privacy and confidentiality of medical information during and after the study. All soft data related to the survey were password-protected, while hard copies of materials were kept under lock and key. The study followed the Helsinki Declaration on human and animal research guidelines. Data were entered into an Excel spreadsheet, and analysis was conducted using the IBM Statistical Package for Social Sciences Version 24.0 for Windows (IBM Corp., Armonk, NY, USA). Percentages (%) and frequencies (n) were used to report categorical variables, with continuous variables being reported as means. We used simple descriptive statistics to summarize the participants' socio-demographic characteristics and bivariate analysis to examine the associations between background characteristics and alcohol and psychoactive drug use. We used multivariate logistic regression models to explore the demographic correlates of alcohol and psychoactive drug use. A p-value of <0.05 was considered to be statistically significant. We tested for collinearity using variance inflation factors (VIF), and none of the variables had a VIF of more than 2, indicating that there was no multicollinearity. Most of the participants were male (59.5%), single (74.9%), Black (87.2%), aged 35 years and above (68.2%), and had attained an education level of Grade 8 to 12 (57.2%). The majority of the participants were unemployed (81%). However, 67.4% (n = 263) of the total participants received social assistance from the government in the form of disability, pensions, and child-care grants. Approximately 47% of the participants (n = 183) reported a combined household income of R2001-R5000 per month. Most of the participants lived in urban areas (83.6%) with their families (84.9%) ( Table 1) . Schizophrenia was the predominant mental illness diagnosed among the participants (40.8%), followed by cannabis-related disorders (19.2%) and major depressive illness (18.5%) (Table S1 ). Of the total participants (N = 390), 64.4% and 33.3% reported ever-use and past-year use of alcohol, respectively ( Table 2 ). The prevalence of ever-use and past-year use of alcohol was higher among males (77.2% and 37.9%, respectively) compared to females (45.6% and 26.6%, respectively). The prevalence of past-year and ever use of alcohol reduced with an increase in age. In the unadjusted logistic regression model, male sex, younger age (<56 years), source of income, and having a lower level of education (