Assessment of depression and its associated factors among patients with type 2 diabetes

Background: Diabetes mellitus (DM) patients have an increased risk of developing several serious health prob-lems in comparison with patients without DM. Depression is one of the more common co-morbid conditions found among diabetic patients. It is associated with poorer glycaemic control, reduced compliance with treatment, and increased complications, which can lead to impaired quality of life. Objective: This study was designed to assess the prevalence of depression and to recognize its associated factors among patients with type 2 diabetes mellitus attending a tertiary care hospital in Dhaka, Bangladesh. Methods: This cross-sectional study was conducted from July 2019 to February 2020. We recruited 318 types 2 diabetics (T2DM) patients who attended an out-patient diabetic clinic at a tertiary care hospital in Dhaka, Bangladesh. We used Patient Health Questionnaire-9 (PHQ-9) to assess depression among the subjects. A semi-structured questionnaire was used to collect demographic, clinical, and diabe-tes-related information after informed written consent was taken from the patients.


Introduction
Two major non-communicable diseases which have become global epidemics and cause significant mortality and morbidity are diabetes and depression (1). International Diabetes Federation (IDF) also declares diabetes as one of the largest global health emergencies of the 21st century (2). This indicates that the prevalence of diabetes mellitus has reached epidemic levels globally (3). People with poorly controlled diabetes have an increased risk of developing several serious health problems. In many countries, diabetes is a leading cause of cardiovascular disease, blindness, kidney failure, and lower limb amputation (4).
Depression is a common and serious medical condition with a lifetime prevalence ranging from approximately 11% in low-income countries to 15% in high-income countries (5). Like any other chronic illness diabetes and depression also have a negative impact on its sufferer. People with diabetes and co-morbid depression can have poor self-care, treatment adherence, and glycaemic control as well as increased morbidity and mortality (6). Evidence also suggests that the prevalence of depression is moderately increased not only in prediabetic patients but also in undiagnosed diabetic patients, and markedly increased in the previously diagnosed diabetic patients compared to normal glucose level individuals (7). The prevalence rate of depression is three times higher in patients with type 1 diabetes and two times higher in the type 2 diabetes population compared with the general population worldwide (8). Diabetes and depression can have a bidirectional relationship. Due to the chronic nature of the disease and numerous complications, patients with diabetes tend to become depressed while di-abetes can appear in depression due to an increase in counterregulatory hormones (9). The prevalence of diabetes, as well as depression, is increasing in Southeast Asia (10). According to different studies reports, the prevalence of depression among patients with diabetes in Bangladesh is ranging from 15.3 to 36% (11,12). Diabetes is more prevalent in urban than in rural population 3 and frequent exposure to several unfavorable conditions make them vulnerable to depression. In the present study, it was speculated that persons with diabetes would have a higher prevalence of depression. Data on depression in the urban-type 2 diabetic population of Bangladesh is still inadequate. Therefore, this study was designed to explore the prevalence of depression and to recognize its associated factors influencing depression among patients with type 2 diabetes mellitus attending an out-patient clinic at a tertiary care hospital in Dhaka, Bangladesh.

Study design & Study setting
This single center-based descriptive type of cross-sectional study was carried out among 318 types 2 diabetics (T2DM) patients attending a diabetic clinic at MARKS medical college hospital in Dhaka, Bangladesh from July 2019 to February 2020. This is a multidiscipline hospital and serves a good number of diabetes patients in the north part of Dhaka city.

Inclusion and Exclusion criteria
We purposively enrolled a total of 318 types 2 diabetes patients aged over 30 years attending the diabetic clinic. Informed written consent was taken from all. All the subjects were diagnosed with type 2 This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited (CC BY 4.0) diabetes mellitus for at least 6 months. Patients with a previous history of psychiatric illness or currently treated with psychiatric medication and pregnant patients were excluded from the study.

Ethical consideration
The ethical approval was obtained from the ethics review board of the hospital.

Data collection, Study instruments, and Measurement
The study was approved by the proper authority of the Institutional Review Board. Two questionnaires were used for data collection. A semi-structured questionnaire was used by the interviewer for the collection of socio-demographic information, anthropometric, clinical, and diabetes-related variables. Depression was assessed by Patient Health Questionnaire 9 (PHQ-9) and relevant data were collected by a self-administered questionnaire of the local language (Bengali). The questionnaires were pre-tested in a similar setting on 20 type 2 diabetes patients in the outpatient department of MARKS medical college hospital. Feedback from the field testing was used to improve the contents of the questionnaire. The diagnostic validity of the PHQ-9 was established in studies involving primary care (13). They were designed by the investigator and validated by a clinical psychologist, among other experts. It is considered a reliable tool for the diagnosis of depression developed by Kroenke et al. (13). The PHQ-9 cut-off score of ≥10 had a sensitivity of 88% and a specificity of 88% to diagnose major depression (14). The reliability and validity of the tool have indicated it has sound psychometric properties (13).
The questionnaire is free to users and available in Bengali and over 29 other languages other than English. The Bengali version of the PHQ-9 questionnaire was evaluated previously and used in the different studies for diagnosis of depression 12. Minimal (or no depression), mild, moderate, moderately severe, and severe depression are defined by the tool as total scores of 0-4, 5-9, 10-14, 15-19, and ≥20, respectively (13). Depending on the PHQ-9 scoring system subjects are again categorized into two other groups, i.e. scores≥5 and ≥10. Score ≥5 includes all those subjects who had depression of any categories from mild to severe (i.e. mild, moderate, moderately severe, and severe depression). And score ≥10 categories those who had depression of moderate to severe categories (i.e. moderate, moderately severe, and severe depression) (14). The questionnaire comprised information about socio-demographic factors such as age, sex, residence (rural or urban), marital status (married, unmarried, widow), education (primary, secondary school, higher secondary school, and graduation), occupation (service holders, businessman, homemaker and retired), monthly income in Bangladeshi taka (BDT) [according to Asian development bank categories15: upper (≥ 100000 BDT), middle (25000-50000 BDT) and low (<5000 BDT)]. The questionnaire also included health-related variables i.e. duration of diabetes, height, weight, body mass index (BMI), blood pressure, fasting blood sugar (FBS), postprandial blood sugar (PPBS), glycosylated hemoglobin (HbA1c) level, presence of other co-morbidities and diabetes-related complications (eye, cardiovascular diseases, cerebrovascular diseases, kidney diseases, peripheral vascular diseases, etc).

Statistical analysis
Data were analyzed with Statistical Package for Social Science (SPSS) software version 20. The means and standard deviations were used to describe continuous data. Categorical variables were estimated by frequencies and percentages. Association between depression and categorical variables was calculated using the Chi-square test. p-value <0.05 was considered as significant.

Baseline characteristics of the study participants
The study recruited 318 types 2 diabetes subjects with a male to female ratio of 28:72. The mean age of the study subjects was (±SD) of 48.88 ± 11.51 years. Among them, 8.5% of the study population was over the age of 66 years. The majority of the subjects (84.0%) were married. More than 60% of the subjects completed secondary or higher education. About half of the subjects were housewives but unemployed, and one-third were service holders or businessmen. Among all, 39.0% of the study population had low income. Anthropometric and socio-demographic data of the subjects are synopsized in Tables 1 and 2.  About two-thirds (62.6%) of study participants reported having other co-morbidities (e.g. Bronchial asthma, osteoarthritis, hypertension, dyslipidemia, etc.); (p=0.66). The presence of diabetes-associated complications (macro and microvascular) was significantly higher among female than male patients (49.4% vs. 18.9%); (p=0.84) ( Table 3).

Discussion
In our study, the overall prevalence of depression was higher (64.1%) in comparison to previous studies done in Bangladesh (11,12). This figure was also much higher in comparison to the prevalence of depression among the adult general population of Bangladesh (4.6%) (17). However, this finding is comparable to those reported in many hospital-based studies of different countries ranging from 21% to 83% with the majority having a prevalence of 41% and above) (18).
In the present study, depression was more common among females (46.2%) than male subjects (17.9%). Similar reports were appeared in many other studies (18,19,20). The majority (72%) of our study subjects were female. Among them, most were housewives and unemployed. We found moderate to severe depression score (≥10) was high among housewives and low among businessmen. Another study (21) also reported a significant correlation between unemployment status and depression score.
Advancing age is usually a predictor for depression and more so in patients with diabetes (22). A significant association between age and depression was also noticed in our study. Reports from different studies advocate that the civil status of subjects with those being married has fewer chances of depression (23). This has been observed in our subjects as well. Depression was markedly lower among married subjects in contrast to those who were widowed.
Depression was deemed to be linked with lower education levels in our study. Association of education with depression was also revealed by some other studies (24,25). Individual monthly income was also seen to be correlated with depression status in our study. Similar findings were also reported in some other studies (26,27). Depression is more frequent in rural residents compared to urban counterparts reported by other studies (28.29). But in our study, the proportion was higher among the urban than rural diabetic population; as the current study was done in an urban tertiary care hospital.
We found higher BMI was significantly associated with a substantial risk for moderate to severe depression. Many studies also have noted a similar relationship between obesity and depression (30). In our study, patients with a longer duration of diabetes (≥10 years) had a high depression score. But some other studies did not find any correlation between the duration of diabetes and depression (21). Diabetes associated complication i.e. macro or microvascular is well studied to have adverse psychological effects among patients with diabetes (31). Findings of some recent studies exhibit that the risk of depression is significantly correlated with the number of macro or microvascular complications (32,33). Our findings also approved the association of diabetic complications (macro/ microvascular) with depression. Contrarily, the presence of other chronic diseases such as hypertension, dyslipidemia, osteoarthritis, asthma, etc was also significantly associated with moderate to severe depression. These reports match with findings of many other studies (20,28). The negative influence of depression on quality of life and glycaemic control is well known (33). Poor glycaemic control was spotted as a significant forecaster of depression in our study. Several other cross-sectional studies also support our findings and reveal the association of depression with poor glycaemic control (34).

Conclusions
In our study, the prevalence of depression specifically moderate to severe is very high among adults with type 2 diabetes. We found depression was significantly associated with older age, female gender, widow, residing in urban areas, low income, lower educational level, unemployment status, high body mass index, diabetic complication and/or other co-morbidities, and longer duration of diabetes.
The findings indicate that patients with diabetes should be inquired for depression or mental disorders and it is also important to incorporate mental health care into the management of diabetes.

Limitation of the study
This is a PHQ-9 questionnaire-based study. The cross-sectional nature of this study could not give any tangible perception into the causal effect relationship. Inclination to recall bias is one of the lim-itations of our study. Secondly, self-reported by patients and previous medical record book-based data about complications or co-morbidities could not be verified by clinical or laboratory investigations. A control or comparison group could be used as a testimonial in the precise estimation of depression among diabetic and non-diabetic individuals.