In this study, we explored the prevalence of comorbid substance used disorders and associated factors of psychiatric patients in Dessie referral hospital by using an institution based cross-sectional study design. This is a preliminary study as a baseline for the project of harm reduction strategies in the above mentioned hospital. The study was conducted at the Dessie referral hospital from November 1/2017 to December 28/2017. Dessie referral hospital is located in South Wollo, Dessie town which is found 401 Kms north east of Addis Ababa. The hospital provides services for greater than 7 million populations. It has around 16 departments. Among these departments, psychiatry is one which has inpatient services together with other medically ill patients and outpatient mental health services for approximately 1200 outpatients per month.

Three hundred thirty six adult psychiatric patients that attended to Dessie referral hospital psychiatry department were included in this study. Psychiatric patients that were severely ill and unable to communicate with different reasons for data collection periods were excluded. A Systematic random sampling technique was used to select 336 adult psychiatric patients.

A structured questionnaire was first being developed in English and translated into Amharic then back-translated independently into English to check for consistency and semantic validity. Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST): Developed by an international research team funded by WHO in 1997 was used to assess to comorbid substance used among psychiatric patients with some modifications. We substituted khat instead of amphetamine since khat is an amphetamine like substance and amphetamine is not available and used in the study area. As the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) was developed mainly for drug use, but can be used for other substances, including alcohol and tobacco as well, particularly in high prevalence settings, it is being considered as an instrument of choice when the goal is to address a range of different psychoactive substances. It is a brief screening questionnaire (8 items) developed for primary care which helps health care providers to identify patients that may have hazardous, harmful or dependent use of one or more substances. It was validated in many cultures and languages. The score is different to specific substance and the risk level is classified as low, moderate and high. Since the moderate and high risk levels need intervention, we dichotomize the findings into two; having risks and has no risk and also for analysis purpose.

Data were collected from study participants using pre-tested written, structured questionnaire by using face to face interview technique. Data collectors were three psychiatric Nurses. The principal investigators and one trained mental health professional specialist was monitored and follow the whole activities throughout the study period. The responsibility for the data collectors was to read the questionnaire, to write the response of participants and check for the appropriateness of data while still in the study site. The questionnaires were filled after obtaining written consent to the respondents. The collected information about the psychiatric patients was reported to the supervisor every day to enable taking immediate action in case inconsistencies or problems happen to the reported data. The supervisor was providing all items necessary for data collection on each data collection days, checking filled questionnaire for completeness and solving problems that would happen to the data collection process.

The following measures were taken to maintain and increase the reliability and validity of the study. The questionnaire was pre tested on 5% of study participants that were done on psychiatric patients that have been following up at Selam general hospital and modifications were incorporated to the questionnaire as needed. The supervisor and data collectors were trained in one day about the instruments and data collection techniques. Follow up of activities by the principal investigators and the supervisor was done during the phase of data collection. Adequate related literatures were reviewed; opinions and comments from experts were obtained throughout the research process to ensure and maintain the external validity of the study. Data were checked and entered by the principal investigators and cleaned before analyses. Epidata version 3.1 double entry was used for data entry. All of the entered data were checked again before final analyses.

Data were checked for its completeness on the day it was collected. Data were entered into a computer using Epidata version 3.1 then transferred to SPSS version-20 for analysis. Then outputs were presented using tables and figure. Descriptive analysis, such as frequencies, proportions, median and means were used. Presence and degree of association between comorbid substance used disorders and the respondents' socio- demographic features and other variables were investigated by binary logistic regression and finally all independent variables whose p- value ≤ 0.25 was entered into multivariate analysis to identify independently associated factors for comorbid substance used disorder. A p- valued < 0.05 was considered as statically significant.

Ethical clearance was obtained from the ethical review committee of the Wollo University college of Medicine and Health Science. An official letter of co-operation was also written to Dessie referral hospital and specifically to psychiatry department. Written informed consent was obtained from each study participant. Individuals who do not volunteer to continue from the beginning or from any part of the interview were respected right to do so. Privacy and strict confidentiality were maintained during the interview process as well as anonymity was kept during data processing and report writing.

A total of 336 psychiatric patients were interviewed with a response rate of 94.6 %. Among all, the majority 181 (53.9%) were males and their age was ranged from 18-90 year with a mean age of 34.1 ± 13.2 years, and Majority 294 (87.5%) were Amhara by ethnicity. Most of the respondents, 236 (70.2%), was followers of Islamic religion followed by 89 (26.5%) followers of Orthodox religion. Relatively majority of the respondents 153 (45.5%) were married followed by single 138 (41.1%). With regard to educational status, higher proportion 100 (29.8%) of the respondents had attended primary school (1-8) and nearly the same proportion of the respondents 95 (28.3%) were illiterate. Among the total respondents, one thirds 114 (33.9%) was unemployed. For more information, see table 1. The median family monthly income of respondents was 900.00 Ethiopian Birr and the range was too high (0-50,000Ethiopian Birr). Sixteen (4.8%) of respondents had no any income and they were living by begging and support from different sources. Out of the total respondents, majority 156 (46.4%) was living with their spouse and children followed by 131 (39.0%) living with their parents.

The majority of patient’s diagnosis 158 (47.0%) was schizophrenia followed by major depressive disorder 82 (24.4%). In line with the patient’s diagnosis, high proportion 139 (41.4%) of them was on antipsychotic medication followed by 91 (27.1%) antipsychotics and antidepressant medications. For majority 234 (69.6%) of the patients, the onset of the illness was between 15 and 34 years of age, the most productive age. The mean ages of onset of illness were 29.88 ± 13.1 years. Regarding the duration of the illness, one third of the patients 117 (34.8%) had less than 2 years, followed by 103 (30.7%) 5-10 years. The range of the duration of illness was 1 month to 44 years and the median years of duration of illness were 4.0 ± 4.86 years. The majority of the patients 295 (87.8%) had no history of admission to hospital and among those who had admission, 12 (29.3%) were admitted two times and above. The majority of the respondents 253 (75.3%) had no family history of mental illness as shown in table 2. Of all respondents, 202 (60.1%) had a family history of psychoactive substance used habitually. Among those who had been having follow up before, 140 (41.7%) received psycho-education about substance use problems.

Table 1: Socio-demographic characteristics of the respondents at Dessie referral hospital psychiatry clinic, 2017 (n=336).

Table 2: The illness related conditions of the psychiatric patients at Dessie referral hospital psychiatry clinic, Dessie town, 2017 (n=336).