Prevalence and Associated Factors of Depression in Outpatients of Internal Medicine Department of Kigali University Teaching Hospital, Rwanda

in ill than in the is not depression rates in medicine outpatients consulting the in with to it is co-morbidity in this it has a on their and In the absence of in our we this our with in might be associated with depressive

Depression is a common mental disorder that is encountered in all clinical facilities (1). It is the leading cause of disability and a major contributor to disease burden worldwide. The global prevalence of depression has been increasing in recent past years (2). The prevalence varies by age, peaking in older adults (above 7.5% among females and above 5.5% among males aged 55 -74 years).
Depression also occurs in children below the age of 15 years, but at a lower rate than older people (3). The consequences of depression are shown by increased medical expenses, dissatisfaction with medical services and with doctor-patient relationship, increased level of disability, negative impact on disease outcome, impact on family-broadly the social costs but most importantly the individual's level of suffering (4,5). Depression was shown to have an impact on co-morbid health conditions; it increases the total health expenses of chronic medical diseases among older adults(6). Physical health conditions increase the risk of suicide (7). Low-income level has been associated with depression(8).
About 50% of patients with depression attend general hospitals due to the somatic expression of their disease. Headache, dizziness, palpitation, weakness, abdominal discomfort, joint pain, burning, and tingling sensations, and vague pain are the common somatic complaints of depressed patients which made them more likely to see neurologists, gastroenterologists, and cardiologists (9,10). The high prevalence of depression in separated or divorced individuals is due to both the high proportion of depressed people whose marriages become disrupted, and also the high proportion of non-depressed people whose marital disruption leads to depression (11). In a study conducted across European countries, they found that a higher socioeconomic index score acted as a protective factor against depression; income was among the predictor of depression (12). Unemployed adults had three times greater odds of having depression compared with employed adults (13). Personal and family history of depression was associated with the prevalence of the diseases (14). Depression is also associated with a more recurrent course, worse impairment, and greater service utilization. People with biological relatives two previous generations affected with major depressive disorder were at the highest risk of depression (15).
People with medical illness have higher rates of depression, it is an often unrecognized co-morbidity in this group, and it has a major impact on their function and disability. In the absence of studies in our region, we did this research in our country, with interest in what factors might be associated with depressive symptoms in our setting.

Methods
This is a descriptive cross-sectional study conducted in the internal medicine outpatient department of Kigali University Teaching hospital (CHUK) from October 7 to November 6, 2019. The University teaching hospital of Kigali/CHUK is the largest hospital located in the District of Nyarugenge at KN 4 Ave, Kigali City. It is also the biggest referral hospital in the country with a capacity of 519 beds. CHUK provides quality healthcare to the population, training, clinical research, and technical support to district hospitals. The hospital is organized into 'divisions'. One of these covers clinical areas such as the clinical service division and other is non-medical -for example, corporate service and research & education divisions.
Each clinical division is managed by a team -usually a consultant and a senior nurse. Non-clinical divisions are led by executive directors. Every division has its own budget and reports to the hospital's management team or Hospital Administrative Board. The divisions are split into medical, nursing, allied directorates and up to the clinical departments and units. CHUK clinical service division offers varied health care services such as surgical services, which include pediatrics, urology, orthopedic surgery, general surgery, plastic surgery, neurosurgery, ENT surgery, ophthalmology and dentistry, specialized clinics in the internal medicine department: nephrology, cardiology, Pneumology, dermatology, oncology, and endocrinology. Nursing care and Allied healthcare services. The internal medicine department has an outpatient department (OPD) and Inpatient service, the OPD receives monthly at least 1200 patients with regular appointments or new consultations.
Participants, Sampling, and Tools: The participants were the outpatients who consulted the department during the prede ned period. The sample size was calculated according to Slovin's formula n = N / (1+ Ne 2 ) where sample size n (300) was taken from the given population N (1200) with the largest margin error of e (0.050). The patients were approached for participation in the study while they were waiting for their appointments with their physicians (Figure 1) below Patients aged 18 years and older on the day of the appointment and who agreed to participate in the study after written informed consent were included. We excluded patients if severe illness, special needs patients below 18 years, and patients with cognitive impairment.
The data were collected by using a structured questionnaire including the Patient Health Questionnaire-9 (PHQ-9). The questionnaire collected socio-demographic and clinical characteristics; it was translated into the local language (mother tongue) and accepted by the national mental division and Department of Psychiatry of the University in the previous studies regarding the screening of depression.
The clinical questions were aggregated in four domains such as family psychiatric history, personal psychiatric history, chronic medical illness, and current chief complaints. The score of 10 was used as the cut-off score for diagnosing likely depressive disorder(16), no depression below 10, and three types of depression above 10 according to severity (Table 1) below. The questionnaire was completed by the patient or researcher depending on the choice of the participant. The data collected were entered into statistical package for social science (SPSS) version 16 for analysis purposes. We carried out descriptive statistics for all variables. Chi-Square test was calculated to determine associations between variables and a 2-sided P value of <0.05 was considered statistically signi cant Bene ts and signi cance The participants were offered information about their level of depression and ability to get biopsychosocial support the same day as those ones who scored 10 or more were referred to the mental department.
The study results could be used to give a valid recommendation to the ministry of health through the hospital in integrating the systematic screening for depression among the adult outpatients attending the internal medicine. It could also remind and motivate physicians to screen patients for depression and other mental illnesses. The results could help the policymakers in upgrading the existing policy about mental disorders. As it is the rst study, it could serve as baseline data for future studies.

Results
Three hundred patients aged between 20 and 96 years were included in the study with 49% aged 45 years and below (  Considering suicidal ideas in relation with socio-demographics and clinical factors (Table 3 below), the signi cance was found in socio-economic status with a diminution of suicidal ideas while improvement of socioeconomic status. reported by Afolabi and his colleagues in Nigeria among the patients attending the family practice clinic (19) The differences in the prevalence rates in various studies considering the present study could be attributed to the variation of demographic factors medically-related factors of the studied population, particularly age, health conditions of the participants, and use of different tools to diagnose or screen depression in various studies as well as cultural differences

Predictors of depression
In the present study, there was a signi cant association of depression and chief somatic complaints (P=0.026) more patients who had neurological disorders were more likely to have depression at a rate superior to or equal to 50%. The review by Rickards had shown the relationship between depression and neurological disorders where it is common in epilepsy, Parkinson's, multiple sclerosis, and brain injuries (20). Headache can be a cause or complication of depression. Fourteen percent of the study patients presented with headache as a chief complaint; of those 66.7% had depression. This rate was higher compared to 25.4% and 32% reported respectively in previous studies done by Maeno et al. 2007 and Kegowicz and Starkey 2009 (21,22).In the present study, participants who reported low socioeconomic status and marital status were not presenting a signi cant relationship statistically, which contrasts with other researchers' ndings (23,24).
The prevalence of depression in patients with hypertension was 34.7%. This nding was higher than the (21.3%) observed in a systemic review and meta-analysis by Li et al. 2015(25). The relationship of depression in patients with no chronic diseases versus any types of chronic diseases was not statistically signi cant (P=0.547). Co-morbid hypertension and diabetes in patients were more strongly associated with depression than hypertension alone, where patients with such co-morbidity presented a prevalence rate greater than 53.1%. This percentage agrees with what Alkhatami et al. 2017 reported regarding the prevalence of depression among patients with hypertension and diabetes at primary health care (57.3%) (26).
Family history and psychiatric history were not predictors of depression among study participants in this study. There was no statistical difference between patients with no history versus any type of history respectively in accordance to P=0.385 and P=0.194. This is also in contrast with other studies' results (14,15). The lack of association could be explained by the question asked in the interview which was focusing on past history of being treated for depression. It might also be due to the low numbers who report a past history.
In the present study, patients' gender was not a predictor of depression (P=0.181). In contrast to our ndings, the gender differences in depression were reported in a study done by Ford and Erlinger,2004 in which women were found to have a higher prevalence rate of depression than men (27).
In the present research, socioeconomic indicators such as employment and living alone were not associated with the development of depression. There is a statistical difference between patients with no education versus any level of education(P <0.001), Bjelland et al. supported the idea that higher educational level may protect against depression but the mechanisms of protection may include other factors such as personal characteristics related to levels of resilience to stress, the level of stress exposure and somatic health (23).
In this study, age was not a predictor for depression and there was no statistical signi cance between patients younger and older than 45 years of age. Aging was associated with the occurrence of physical disorders but it was shown to be independent of depression(28). Our study found a frequency of depression as 35.5% of patients who were living alone compared to 46.8% among patients who were living with others. This contradicts a study by Sthal where they observed elevated depression symptoms among people are living alone (29).
As many as 22.7% of patients were presenting with Suicidal ideation, there was a strong relationship between low socioeconomic status and suicidal ideas, therefore it would have been better to complete the screening with the suicide severity rating scale in order to identify the patients who might need immediate treatment

Strengths and limitations
This study was limited to one academic centre and its ndings cannot automatically be generalized to other types of health facilities. Patients attending internal medicine with severe illness were excluded from this study but from the trend seen in chief complaints, it is likely that the depression rate among these patients will be higher than the patients coming for follow-up visits. PHQ-9 is a useful selfinstrument for screening and long-term follow-up of depression. However, this study is a cross-sectional and causal relationship can't be attributed

Conclusion And Recommendation
Prevalence of depression among internal medicine outpatients at a general tertiary hospital was high.
The use of a depression screening instrument like PHQ-9 with the objective of improving early detection and treatment of this mental illness should be highly encouraged. Integrating mental health into chronic diseases management should be implemented as a priority to reduce the mental health gap and improve chronic diseases outcomes.
List Of Symbols And Acronyms  Prevalence of depression among outpatients attending internal medicine department Presenting depressive symptoms among study patients

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Datasheet.xlsx