Study design, period, setting, and population
An institutional-based cross-sectional study was conducted from April to May 2018 in Zewditu Memorial Hospital which is found in Addis Ababa, Ethiopia. Center for Disease Control and prevention (CDC-Ethiopia) helped the launch of Ethiopia’s first antiretroviral therapy (ART) program at Zewditu Memorial Hospital in July 2003, and in March 2005. The program received technical assistance from Johns Hopkins University’s TSEHAI Program. Zewditu Memorial Hospital became the largest HIV clinic in Ethiopia and a leading hospital in the treatment of antiretroviral therapy patients. Currently, the hospital treats over 7299 patients each month. There were 17,857 HIV-positive patients having HIV care follow-up in the hospital in 2018. The study population was patients who were attending at ART clinics during the data collection period. All adults who were seriously ill and unable to communicate were not included in the study.
Sample size and sampling technique
Sample size (n) was calculated based on single population proportion formula, by assuming 95% confidence level, the prevalence of poor sleep quality among HIV/AIDS patients which was found to be 59.3% in Nigeria  and a precision of 5% between the sample and the parameter was taken. α =0.05(95%) =1.96
By considering a 10% non-response rate the final sample size was 408. We used a systematic sampling technique to select the four hundred eight (408) HIV/AIDS patients who were included in our survey. We determined the sampling interval by dividing the total study population who had to follow up during the average 1-month data collection period (3264) by total sample size (408). Hence, the sample interval is eight. We selected the first study participant by lottery method and the next study participants were chosen every 8th interval for interview.
Data collection tools and procedures
Data were collected by trained nurses by face-to-face interviewing of the participant attending ART service using paper and pencil. The questionnaire was pre-tested by taking 5% of the calculated sample size and modified accordingly to easily understand by the study participants. The questionnaire contained socio-demographic characteristics (age, income, education, occupation, marital status, and others), questions to collect data on clinical factors and standard tools to address other independent variables of the study.
Data on the components of sleep quality was collected by using the Pittsburgh Sleep Quality Index (PSQI), a self-report measure instrument composed of 19 items evaluating seven components of sleep. Each part was scored (range: 0-3; higher scores indicating worse sleep). A total global PSQI was derived by summing the seven components (range: 0 to 21; higher scores indicating poor sleep quality). A global PSQI score >5 yielded a diagnostic sensitivity of 89% and specificity of 86.5 %( к=0.75, p≤0.001) in distinguishing “good” from “poor” sleepers. “Good sleep” was defined as global PSQI scores of 0-5 and “poor sleep” was global PSQI scores of 6-12. The tool was validated in Ethiopia among community dwellers having Cronbach’s alpha of 0.59, sensitivity of 82%, and specificity of 56.2% .
Hospital anxiety and depression scale (HADS) were used to assess anxiety and depression. The tool has anxiety subscale (HADS-A) and the depression subscale (HADS-D). It has a cutoff point ≥ 8 for each subscale . HADS had an internal consistency of 0.78 for the anxiety, 0.76 for depression subscales and 0.87 for the full scale in a validation study conducted among HIV Infected Patients in Ethiopia . HIV/AIDS related stigma scale was used to assess stigma. It is a 12-item screening tool developed by Annelies Van Rie, Sohini Sengupta. The tool has 4-point Likert response. Each items were scored with 0 (strongly disagree) and 3 (strongly agree). Participants who scored above the mean score were considered as stigmatized[24, 25]. The scale demonstrated acceptable internal consistency (Cronbach’s alpha=0.73) in a study conducted in Uganda.
Sleep hygiene index (SHI), a 13-item self-report measure designed to assess the practice of sleep hygiene behaviors. Each item is rated on a five-point scale ranging from 0 (never) to 4 (always). Total scores range from 0 to 52 with a higher score representing poor sleep hygiene. Oslo Social Support Scale (OSS) was used to assess respondents support system which is a 3 items measure of support with internal consistency (Cronbach’s α = 0.50) .
Data quality control issues
Training was given to the data collectors and supervisors on the data collection tool and sampling techniques. Supervision was held regularly during the data collection period both by the researcher, co-investigators and supervisors to check on a daily basis for completeness and consistency.
Data were analyzed using SPSS version 20. Description statistics (frequencies, proportions, means, and standard deviations) were used to present the sociodemographic characteristics and the prevalence of sleep quality. Both bivariate and multivariate logistic regression analysis were carried out to see the association of each independent variable with the outcome variable. A p-value of less than 0.05 was considered statistically significant, and an adjusted odds ratio with 95% CI was calculated to determine the association.
Ethical clearance was obtained from the joint ethics committee of the University of Gondar and Amanuel Mental Specialized Hospital and Addis Ababa health bureau based upon Declaration of Helsinki (DoH). The data collectors clearly explained the purpose and importance of the study to each study participant before they proceed into actual activities. Information was collected after obtaining written consent from each participant. Written informed consent was obtained from the study participants and comparison of study participants after oral information about the study, including an assurance that they could withdraw from the study at any time. Confidentiality was maintained by anonymous questionnaire.