The study was a community-based cross-sectional study.

This study was conducted in Kuru, located some 20 km to Jos, the capital of Plateau State. Kuru is a conglomerate of several villages and hamlets within Jos South LGA. The indigenous language spoken in Jos South is Berom but Hausa is generally spoken and is the language of business. The estimated population of Kuru in the last National Population censors of 2006 was 36,054. The predominant occupation of the town’s residents is farming.

The study population was elderly men and women, aged 60 years and above residing in the community, at least not less than 6 months preceding the commencement of the study.

The projected sample size that was required to meet the set objectives at 95% CI was calculated using the formula

N = Z 2 p q / d 2 [12]

where n = minimum sample size, z = standard normal deviation at 95% of the confidence interval, a constant = 1.96, p = proportion of the elderly with lifetime use of alcohol from a previous study which was 45.2% = 0.452²⁵ (11); q = complementary probability = 1 − p = (1 − 0.452) = 0.548; d = precision level 5% = 0.05. A minimum sample of 381 was calculated, but increased to 419, to allow for a 10% non-response or attrition rate.

All elderly men and women from ages 60 years and above who have resided in the community for more than 6 months were included. Elderly Patients who are too acutely ill to get into meaningful clinical engagement were excluded from the study.

Kuru was purposively selected based on anecdotal reports of high rates of use of alcohol and other psychoactive substances by her residents. However, there was no documented study on the rate of use and its effects in the community.

The estimated population of the elderly 60 years and above in Kuru in the last nationwide censors in 2006 obtained from the office of National Population Commission in Jos was 1530. Due to the limited number of elderly in the community and the exploratory nature of the, a total population sampling of all the elderly individuals in the community was done and all elderly members of the community who met the inclusion criteria and gave consent were recruited into the study until the sample size was achieved. The selection of participants was done from house to house. A house mapping was conducted to identify all the houses where the elderly resides and the total number of such houses was noted.

All houses in the town were enumerated, and each house having an elderly person aged 60 years and above were mapped. Using the above process, a total of 422 elderly community dwellers were identified. Out of this number, 16 declined consent and 3 were too ill to respond while 3 others gave very scanty information that their responses could not be used.

Data was collected using a composite questionnaire and was interviewer administered. The data collection instrument was translated to Hausa language and back translated to the English language by two professionals (a linguist and a psychiatrist) fluent in English and Hausa, so as to ensure content validity of the tool.

Socio-demographic questionnaire capturing gender, age, education, income (personal and family), work/occupation (including current regular professional work, other occupations in retirement, non-professional work, voluntary), ethnicity, religion, marital status, living status.

· Social Network (existence of a person who participant can count on in difficult moments);

· Lifestyle, physical activity, self-reported health status;

· History of chronic physical or mental illness;

· Functional status: assessed by a five-item scale eliciting basic household independency, ability to take medicines, bathe/comb hair/dress, ability to eat without support, and physical functioning—walk, sit, lie down, get around indoors, and go upstairs;

· Use of health service: number of doctor visits in the last 12 months and/or hospitalizations in the last 12 months.

General Health Questionnaire 12-item The GHQ-12 was chosen because of its brevity and ease of completion. The validity of the GHQ-12 has been reported to be comparable with that of the longer versions of the GHQ in the identification of minor psychiatric disorders. For the purpose of this study, the GHQ scoring method (0-0-1-1) was chosen over the simple Likert scale of 0-1-2-3, as this particular method is believed to help eliminate any biases which might result from the respondents who tend to choose responses 1 and 4 or 2 and 3, respectively [13]. The GHQ-12 had been previously validated [14] and extensively used in Nigeria [15] [16] [17].

The Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST): The ASSIST was developed by the World Health Organization for screening in primary and general medical care settings and has since been considered as an instrument of choice when the goal is to address a range of different psychoactive substances [18]. This test categorizes respondent’s substance use into low, moderate and high risks on the bases of their scores. With scores of 0 - 10 for alcohol and 0 - 3 for illicit drugs, participants are considered as having a low risk for health and other problems from their use of alcohol and other drugs (AODs), 11 - 26 for alcohol and 4 - 26 for other drugs indicate moderate risks while >26 indicates high risk for health and other problems from their current use of alcohol and other drugs and are likely to be dependent. The ASSIST was used in a previous study [9], and took an average of 3.85 - 6 minutes to administer to the research participants.

Mini International Neuropsychiatric Interview (Alcohol Section): The MINI was used in this study. Responses were rated at the right of each question by circling either “Yes” or “No”. The MINI is a short structured clinical interview that helps researchers to make diagnoses of psychiatric disorders according to DSM-IV or ICD-10. The MINI was designed for epidemiological studies and clinical trials and is divided into modules identified by letters, each corresponding to a diagnostic category [19]. At the beginning of each diagnostic module, screening question (s) are made to the corresponding main criteria of the disorder, after which diagnostic box(es) permit the clinician to indicate whether the diagnostic criteria are met. This module took approximately 1 - 2 minutes to administer in this study.

The ASSIST Feedback Report Card was used in this study to provide substance-specific involvement scores based on calculated standard ASSIST scoring procedures. The scores ranged from 0 - 10 = low, 11 - 26 = moderate, and 27+ = high. Research participants within the low score range on the ASSIST are interpreted as being at “low” risk of health and other problems related to substance use. Finally, participants whose scores fall within the “moderate-high” range will be evaluated as being at “moderate-high” risk of experiencing severe problems as a result of their current use and are likely dependent on the psychoactive substance.

Data was collected from 17th to 31st August, 2018. The data collection team comprised of total of eight [8] trained research assistants; psychologists, social workers and interns. They received training in the content and administration of the data collection instruments as well as the objectives of the study and more importantly, how to carry out the data collection in a non-judgmental way. The research assistants could speak English, Hausa and Berom. The data collection instruments were pretested among elderly people in a different community. This helped determine average time of administration of the instrument and helped to detect and correct minor ambiguities.

Data was analysed using SPSS version 23 statistical software after imputing into an excel sheet and exporting to the SPSS. The dependent variable was substance use/disorder (present/absent). Independent variables were: socio-demographic characteristics, social network, self-reported health status, psychiatric morbidity and use of health services.

Frequencies and percentages were used to calculate the prevalence of psychoactive substance use/disorders. The Chi-squared test categorical associations with level of significance set at p < 0.05.

Approval for this study was obtained from the Ethics Committee of the Jos University Teaching Hospital (JUTH). Permission was sought from the district head of Kuru after explaining the nature and purpose of the study. Due to limited formal education in most of the cohort, the purpose of the study was read out to participants in the presence of a significant other, and informed verbal consent or thumb-printed signed forms were obtained.

All participants who screened positive on ASSIST and those diagnosed with AUDs on Mini International Neuropsychiatric Interview (MINI) were counseled adequately, given brief intervention or referred to the Psychiatry clinic of the hospital for management as the case may be.

A total of 400 out of a total of 422 elderly community dwellers identified were interviewed.

The sample was predominantly female (n = 242; 60.5%), aged between the 60 - 69 years’ age group (n = 249; 62.3%) with a mean age was 69.19 years. All but one of the respondents were Christians. Only 11 respondents were living alone, while 144 (36%) lived with a spouse or with their children (n = 139; 34.8%). The majority of respondents were married (230, 57.5%) and a total of 157 (39.3%) respondents were widowed.

Over half (59%) had no formal education, and past-month employment status of respondents showing that 111 (39.8%) were unemployed. Majority of the respondents (n = 235, 58.8%) were farmers. Seventy-nine percent of respondents earned between N1000 to N10,000 and only 12.8 % (n = 51) earned above N21,000. See Table 1.

Nearly a third (30.8%) reported their current health status as “bad”. While a majority of participants reported having chronic pain (n = 267, 66.8%), those with cardiovascular (n = 58, 14.5%), respiratory (n = 15, 3.8%), digestive (n = 172, 43%)

conditions were in the minority. Over half screened positive on the GHQ-12 with probable psychiatric morbidity (n = 224, 56.0%). See Table 2.

The lifetime and current prevalence of alcohol use were 69.8% and 45.5% respectively. The prevalence of alcohol use disorder (AUD) in this study was (n = 52, 13.0%) (see Table 3).

AUD was significantly associated with being male (OR = 2.57, 95%CI: 1.35 - 3.77, p < 0.01), but not living alone (p = 0.16). Participants with AUDs were more likely to report cardiovascular (χ² = 0.42, p = 0.52), respiratory (χ² = 0.55, p = 0.46), digestive (χ² = 0.17, p = 0.68), difficulty ambulating (χ² = 0.34, 0.56), probable psychiatric co-morbidity (χ² = 2.35, p = 0.13) and chronic pain conditions (χ² = 0.52, p = 0.47), but the relationships did not attain statistical significance.