Scientific Papers

Magnitude of erectile dysfunction and associated factors among adult diabetic men on follow-up at Goba and Robe hospitals, Bale Zone, South East Ethiopia: hospital-based cross-sectional study | BMC Endocrine Disorders

Study area and period

The study was conducted at Goba and Robe hospitals in Oromia region, Bale zone which is located 435 km to the southeast of Addis Ababa, Ethiopia. Goba Referral Hospital is the only referral hospital in the Bale zone that serves as a teaching hospital for medical and health sciences. Data was collected from March 1 to April 30, 2022 by two trained male nurses.

Study design

Hospital-based cross-sectional study design.

Source population

All diabetic men who visit chronic outpatient departments of Goba and Robe Hospitals.

Study population

All adult diabetic men who fulfill eligibility criteria and have follow-up at Goba and Robe hospitals during the data collection period.

Eligibility criteria

Inclusion criteria

Adult diabetic men aged ≥ 18 years who also had a history of sexual intercourse.

Exclusion criteria

Critical patients with hyperosmotic hyperosmolar state or diabetic ketoacidosis, mentally incompetent, Psychotic patients, Patients with previous pelvic surgery, Patients who had pelvic trauma, Patients who had a spinal injury, Patients taking anti-psychotic medications, patients with any malignancies and patients with benign prostatic hyperplasia were excluded from the study through history taking and physical evaluation.

Sample size determination

The study sample size was determined using epi-info 7 based on assumptions of 95% confidence level, 5% margin of error, the proportion of erectile dysfunction 54.3% [19], and non-response rate of 10%. The maximum sample size was obtained for 54.3% proportion. Taking a non-response rate of 10%, the final sample size is 420.

Sampling technique

A systematic random sampling technique was used to select study subjects. The average number of monthly adult diabetic cases seen in the outpatient department of Goba and Robe hospitals within the preceding six months were taken (from Goba 600 and Robe 460) and divided by proportionally allocated sample size (238 for Goba and 182 for Robe hospitals). When we divide this by the allocated sample size for each hospital the quotient is the interval chosen to select each case. Therefore, the respondents were interviewed every 2 and 3 patients for Goba and Robe hospitals respectively. The medical record number was taken for each patient to avoid duplication.


Dependent variable

Erectile dysfunction is the outcome variable for this study. It was assessed using International Index of Erectile Function (IIEF) questionnaire. This questionnaire consists of five questions, each with Likert scale alternative responses. The greatest possible score is 25, which is the score for people who never had an erection problem and the lowest score is 5. The sum of the ordinal replies to the 5 items determines the IIEF-5 score. The erectile dysfunction was categorized into two depending on the IIEF score. Those who score 22 and above were classified as having no erectile dysfunction and those who score 21 and below were classified as having erectile dysfunction. More specifically, those who have a score of 17–21, 12–16, 8–11, and 5–7 are sub-categorized as having mild, mild to moderate, moderate and severe erectile dysfunction respectively [42].

Independent variables

Socio-demographic and economic characteristics

Age, Place of residence, Marital status, Ethnicity, Religion, Occupation, Educational status, Income.

Wealth index

Principal component analysis was done for common household assets. The requirements for the analysis like large sample size, the ratio of cases to variables, KMO > 0.6, and Bartlett test of sphericity statistical significance(P < 0.05) were checked and satisfied. Accordingly, KMO = 0.601 and Bartlett’s test of sphericity < 0.001. Four variables number of chicken, radio, refrigerator, and wrist watch with eigen value greater than one were extracted and explained 24.29%, 18.17%, 13.5%, and 12.77% respectively. The total wealth index variability explained by principal component analysis was 68.74%.

Treatment and lifestyle

Types of diabetes, Duration of diabetes, Types of treatment, BMI, Waist circumference, Hip circumference, Waist to hip ratio, Smoking, Alcohol, Khat chewing, Physical activity, Diet.

Comorbidities and Complications

Hypertension, Antihypertensive medications, Cardiac disease, Cardiac drugs, Eye disease, Renal disease, Stroke, Peripheral neuropathy.

Mental and psychological factors


Operational definition

PHQ-2 and PHQ-9(patient health questionnaire)

were used to assess variable depression. The PHQ-2 inquiries about the frequency of depressed mood and anhedonia over the past two weeks. It includes the first two items of the PHQ-9. The purpose of the PHQ-2 is to screen for depression in a “first-step” approach. Patients who screen positive or have a score of ≥ 3 should be further evaluated with the PHQ-9 to determine whether they meet the criteria for a depressive disorder [43]. PHQ-2 score ranges from 0 to 6 [44]. PHQ-9 tool contains nine questions with four options of answers. Each score was summed up and labeled accordingly [45]. For this study, score of 10 and above were used to define depression with good sensitivity.

International Physical Activity Questionnaire (IPAQ)

This assesses the types of the intensity of physical activity and sitting time that people do as part of their daily lives are considered to estimate total physical activity in MET-min/week and time spent sitting [46]. The total number of days and time in minutes spent on activities within a week will be multiplied by their corresponding MET vigorous, moderate, and walking activities. The outcome was categorized and named as high (At least 3000 MET), moderate (At least 600 MET but less than 3000 MET) and low (unable to classify as high or moderate) for vigorous, moderate, and walking activities respectively.

Waist circumference

Is a measurement of abdominal obesity and provides independent risk information that is not accounted for by body mass index. It is measured with flexible tape placed on a horizontal plane between anterior superior iliac spine and the lower rib. A normal waist circumference for male should be less than 102 cm [47].

Waist-to-hip-ratio (WHR)

Measures the ratio of waist circumference to hip circumference. It determines how much fat is stored on waist, hips, and buttocks. Even though, the cut off value vary from population to population a WHR of 1 and above increases risk of metabolic syndrome [48].

Data collection tool and procedure

After reviewing different works of literature [41, 42, 49,50,51], a written structured questionnaires were prepared to determine the prevalence of erectile dysfunction and associated factors among adult diabetic men at Goba and Robe hospitals. The English version of the questionnaire was translated into Afan Oromo and Amharic, the local languages spoken in the study area. After the translation, the questionnaire was -translated back to English to check the consistency. Data were collected through interviewer administered questionnaire and physical measurements. The standard tools used for this study were WHO STEPS-wise questionnaires, summary of diabetes self-care activities, international physical activity questionnaire, public health questionnaires, international index of erectile function and Ethiopian demographic and housing survey data. The selection of these hospitals was made based on patient flows and the comprehensive service they provide. The data were collected by interviewing the patients individually by two trained male nurses on working days. The international index of erectile function questionnaires was used to screen the patients for erectile dysfunction. A separate and quiet room was used for data collection. The remaining data were retrieved from the patients’ documents using prepared checklists.

Data quality control

The principal investigator gave training for two male nurses before the data collection period. The data collectors were capable of speaking local languages. The training was given on the objective of the study, actual data collection procedure, the contents of the questionnaires, and matters related to confidentiality. Additionally, to maintain the quality of data pretest was conducted by the principal investigator at Ginnir hospital on 5% of the sample size to check the validity and know whether the questionnaires are understandable or not a week before beginning the actual data collection period. The modification was made to the questionnaire that is found with any ambiguity and affects its consistency. During data collection, the principal investigator gave onsite technical support and close supervision. Data were checked daily by the principal investigator for completeness, and consistency and then errors were corrected accordingly before data processing and analysis.

Data processing and analysis

Data were edited, coded and entered by Epidata version 4.6 and were exported to SPSS (version 26) for analysis. The data analysis started from a basic description to the identification of potential factors associated with erectile dysfunction. Bivariable and multivariable logistic regression were used to identify the relationship between erectile dysfunction and various factors. The assumptions of the logistic regression model Hosmer-Lemeshow goodness of fit statistics was checked and satisfied. Multivariable logistic regression was used to identify potential confounding variables. Multicollinearity among independent variables were checked using tolerance and variance inflation factor. Normality for independent variables was checked using Shapiro wilk. Principal component analysis was performed for common household assets. Descriptive summaries were computed as simple frequencies, mean, median and standard deviations. All explanatory variables which resulted in p < 0.25 with the outcome variable in the bivariable were entered into a multivariable logistic regression model to identify factors associated with erectile dysfunction. P-Value < 0.05 was considered statistically significant and the adjusted odds ratio with a 95% confidence interval was used to declare association.

Source link