Prevalence of sexual dysfunction and related factors among diabetes mellitus patients in Southwest Ethiopia

A facility-based cross-sectional study was conducted from July 01–31/ 2018 among diabetes mellitus patients on follow-up at diabetic clinics of Mizan Tepi University Teaching Hospital (MTUTH) and Tepi General Hospital (TGH). MTUTH and TGH are located in Southwest Ethiopia at 585 km and 611 km from Addis Ababa respectively. MTUTH is found in Mizan-Aman town. The hospital provides service for the population that comes from Bench-Maji, Sheka and Kafa Zones and some parts of Gembella regional state. It has a bed capacity of 209 and has over 400 staff members. TGH located in Tepi town, and it provides service for people who come from Shaka, Kafa and Majang Zones.

Adult diabetic patients who had been on follow up at diabetic clinics of MTUTH and TGH were a source population. The study population was a sample of adult DM patients who were on follow up during the study period at the diabetic clinics of MTUTH and TGH, and who fulfilled eligibility criteria. Patients who were mentally impaired and unable to give information and/or had sexual disorder before the onset of diabetes mellitus were excluded. Also, females aged above 50 years, and/or those who had history of pelvic surgery were excluded.

The sample size was determined manually using a single population proportion formula (\( n=\frac{{\left({z}_{\alpha /2}\right)}^2p\left(1-p\right)}{d^2} \)) [14]; based on the assumptions of 95% confidence level, 5% margin of error and a 50% proportion of sexual dysfunction. The prevalence of 50% was taken because there was no similar study done in Ethiopia previously. After adding 10% contingency for non-response, the final sample size for the study was 423. Based on the total number of patients on follow up at each hospital, the sample size was proportionally allocated to the two study settings. Accordingly, 279 samples were allocated to MTUTH and 144 samples to TGH. The study participants were recruited during their regular medical follow-up period consecutively based on their arrival at DM clinics of each Hospital.

Changes in Sexual Functioning Questionnaire-fourteen items (CSFQ-14) was adopted and used to measure sexual dysfunction [15]. The questionnaire to assess socio-demographic and behavioral related factors was developed by reviewing different literature. Also, data extraction checklist was prepared to collect data related to medical history. The tool was translated to a local language, “Amharic” then back to English, to ensure its consistency. Then, it was pre-tested on 5% of the total sample size at Chenna Hospital before the actual data collection to evaluate readability, understandability, completeness, and reliability of the questionnaire, and modified accordingly. Internal consistency for CSFQ-14 was checked and demonstrated Cronbach’s alpha of 0.72 and 0.81 for women and men respectively. Finally, the Amharic version was used to collect the data. Trained personnel collected the data through face-to-face interview method. The most recent recorded information related to medical condition were extracted from patient medical chart. Due to the sensitivity of the issue, female data collectors were assigned to collect data from the female participants whereas male data collectors collected data from male participants.

The dependent variable of the study was sexual dysfunction. The independent variables were socio-demographic information (age, sex, religion, marital status, educational status, ethnicity and place of residence), factors related to medical condition (types of DM, duration of treatment, glycemic control, complication or co-morbidity and depression) and behavioral factors (physical activity, adherence to antidiabetic medication, alcohol, tobacco and khat use).

Sexual dysfunction was measured using CSFQ-14, a standardized tool used to screen sexual activities. The CSFQ-14 has 14 separate items for males and females to assess sexual function. All items were answered on five Likert scales. The responses were summed up to give a total score of 14–70. The score at or below 47 for males and 41 for females indicates the presence of global sexual dysfunction. The tool also measures five types of sexual dysfunction (pleasure, desire/frequency, desire/interest, arousal/erection and orgasm/ejaculation). Scores of less than or equal to 6,9,4,3, and 13 indicate sexual desire/frequency, sexual desire /interest, sexual pleasure, sexual arousal/excitement, and sexual orgasm/completion dysfunctions respectively in males. In females, scores at or below 6, 9, 4, 12, and 11 show the presence of sexual desire frequency, sexual desire/interest, sexual pleasure, sexual arousal/excitement, and sexual orgasm/completion dysfunctions respectively.

Glycemic control was assessed using more recent fasting blood sugar (FBS) measurements’ of three months. Patients were considered in good glycemic control range if the average FBS level was ≤130 mg/dL, and poor glycemic control if it was > 130 mg/dL. Depression symptoms in the past 14 days of the survey were measured using validated patient’s health questionnaire-nine (PHQ-9), which is a validated tool [16]. The PHQ- 9 has nine items, each item having 4-point response scales; ‘not at all’ (0), ‘various days’ [1], ‘more than half of the days’ [2] and ‘nearly every day’ [3]. Summing all these possible, the total score range from 0 to 27 points. The score of 0–4 indicates no or minimal depression, 5–9 is mild depression, 10–14 is moderate depression, 15–19 moderately severe depression and 20–27 severe depression. Overall cutoff ≥10 was used to indicate depression. Data related to complication or comorbidity was obtained from patients’ medical charts based on physician diagnosis. If patients had at least one of these conditions; nephropathy, retinopathy, diabetic neuropathy, diabetic foot ulcer, cardiac disease or hypertension they were considered as they had complication or co-morbidity otherwise no. Complication and co-morbidity were merged together because it is very difficult to differentiate them from secondary data if these conditions were occurred secondary to DM or happened as co-morbidity. Duration on anti-diabetic treatment was assessed based on patient self-report of the years elapsed since they started the medication.

Adherence to physical activity was measured based on the participants’ responses. Patients who had been regularly engaged in moderate (like slow walking or dancing) to vigorous-intensity exercise (like fast walking or running) at least 2 days per week were classified as adherent to physical activity. We assessed adherence to anti-diabetic medication in the past seven days of survey and respondents were considered adherent if they took all anti-diabetic medication as per the prescription of a physician. The respondents were categorized as substance users if they reported consumption of any of the following substances in any amount in the past 12 months; had drunk common alcohols in the area such as beer, wine or local alcohols like ‘Areke’, ‘Teji’or ‘Katikala’; smoked tobacco, chewed ‘khat.

Ethical clearance was obtained from the Institutional Review Board (IRB) of Mizan-Tepi University (Ref No: CHS/205/2018). Data were collected after obtaining written informed consent from all respondents. All the information gathered was kept confidential.

Among planned 423 planned samples, 398 (94%) were successfully involved in the study. The mean age of the respondents was 41.76 (±9.00) years and 40.20% were in the age range of 41 to 50 years. Out of the total respondents, 64.3% were male, 35.4% were protestant Christian, 43.2% had attended secondary and above education, 64% were married, and 71.4% of respondents were urban residents (Table 1).

Nearly two-thirds (67.6%) of study participants were engaged in moderate to vigorous physical activity. One hundred fifty-eight (39.7%) respondents had history of substance use in the past 12 months. The substance commonly used was alcohol (20.9%). Seventy-four percent of the respondents adhered to antidiabetic medication in the past 7 days before the interview (Table 2). Out of 398 respondents, 318 (79.9%) were Type 2 Diabetes Mellitus (T2DM) patients. Based on the American Diabetic Association (ADA) guideline recommendation, 87.9% of the patients were in poor control glycaemic range (FBS > 130 mg/dL). The prevalence of depression among study participants was 36.9% (Table 3). Moreover, 39% of patients were experienced complications related to DM and the common complication was hypertension (24.3%) (Fig. 1).

The prevalence of global sexual dysfunction among study participants was 53.3% (95% CI: 48.35–58.25%). Gender wise, the prevalence of global sexual dysfunction was 52% (95% CI: 46–58%) among males and 56.6% (95% CI: 47–64%) among females. The most prevalent dysfunction among men was frequency of sexual desire disorder (60.9%) followed by orgasmic dysfunction (56.6%). Among women, frequency of sexual desire disorder was the most common sexual dysfunction (46.5%), followed by desire (interest) dysfunction (36.6%) (Fig. 2).

In bivariate binary logistic regression analyses, age, marital status, educational status, physical activity, substance use, adherence to anti-DM medication, type of DM, duration of treatment, complication of DM or comorbidity and presence of depression had a p-value of ≤0.25; hence, they were candidates for a multivariable model. However, sex, religion, ethnicity, and duration of treatment were not significant at p-value of ≤0.25; therefore, they were excluded from multivariable analysis. In the multivariable binary logistic model, age of respondents, educational status, marital status, types of diabetes mellitus, complication or comorbidity, the presence of depression and physical activity were significantly associated (p < 0.05) with sexual dysfunction.

Based on adjusted odds ratio, DM patients in the age range of above 41 years had 3.98 times higher odds of sexual dysfunction as compared to those in the age range of less than or equal to 41 years (AOR = 3.98; 95% CI: 2.32–6.85). The odds of sexual dysfunction was 3.2 times higher for patients who did not attend education in reference to those who attended secondary school or above education (AOR = 3.20; 95% CI: 2.32–6.85). However, there was no statistically significant difference in odds of sexual dysfunction between patients who attended and secondary school or above (AOR =0.46; 95%:0.20–1.00). Participants who did not involve in physical activity had 1.62 (AOR = 1.62; 95% CI: 1.47–1.77) times higher odds of sexual dysfunction than the counterpart. With reference to Type 1 Diabetes Miletus (T1DM), the odds of sexual dysfunction among T2DM patients were 4.52 times higher (AOR =4.52: 95% CI: 2.17–9.42). The sexual dysfunction among patients who had complication or comorbidity was 2.05 times (AOR = 2.05; 95% CI: 1.18–3.58) higher when compared to those who didn’t have such history. The odds of sexual dysfunction was 4 times higher for diabetic patients who had symptoms of depression as compared to those who had no depression (AOR = 4.05; 95% CI: 2.32–7.10) (Table 4).

The study addressed the issue that was not well studied in Ethiopia. However, due to the cross-sectional nature of our study, we cannot be sure of the temporal relationship between different factors and sexual dysfunction. Furthermore, the study was done on a culturally sensitive and embarrassing issue, therefore, social desirability bias could not be ruled out. As a result, patients may under-report the problem. In addition, since the study was conducted among patients in hospitals, the finding may not represent patients on follow-up in other settings.