PREVALENCE AND FACTORS ASSOCIATED WITH DEPRESSIVE ILLNESS IN PATIENTS WITH TUBERCULOSIS IN MULAGO HOSPITAL BY DR

PREVALENCE AND FACTORS ASSOCIATED WITH DEPRESSIVE ILLNESS IN PATIENTS WITH TUBERCULOSIS IN MULAGO HOSPITAL
BY
DR. RACHEAL ALINAITWE MBChB (MAK)
DEPARTMENT OF PSYCHIATRY, COLLEGE OF HEALTH SCIENCES
MAKERERE UNIVERSITY

SUPERVISORS
ASSOC. PROF. NOELINE NAKASUJJA MBChB, MMED, PhD
DR. PAUL BANGIRANA BSc, MSc, PhD
DR. CAROLINE BIRUNGI MBChB, MMED

A dissertation submitted in partial fulfillment of the requirements for the award of the degree of Masters of Medicine (Psychiatry) of Makerere University
March 2018

Table of contents
Table of contents 2
List of Tables 4
List of Figures 5
DECLARATION 6
DEDICATION 7
ACKNOWLEDGEMENT 8
OPERATIONAL DEFINITIONS 9
List of Abbreviations 10
ABSTRACT 11
CHAPTER ONE: INTRODUCTION 12
1.0 BACKGROUND 12
1.1 PROBLEM STATEMENT 14
1.2 SIGNIFICANCE 15
1.3 RESEARCH QUESTION 16
1.4 OBJECTIVES 16
1.4.1 GENERAL OBJECTIVE 16
1.4.2 SPECIFIC OBJECTIVES 16
1.5 CONCEPTUAL FRAMEWORK 17
CHAPTER TWO 18
2.0 LITERATURE REVIEW 18
2.0.1 Magnitude of depression among patients with TB 18
2.0.2 Possible mechanisms for the association between TB and depression 19
2.0.3 Factors associated with depression among patients with TB 20
2.0.4 Effect of depression on TB management 22
CHAPTER THREE 24
3.0 METHODS 24
3.0.1 Study design 24
3.0.2 Study setting and site 24
3.0.3 Study participants 24
3.0.4 Selection criteria 25
3.0.5 Sample size estimation 25
3.0.6 Sampling method 25
3.0.7 Study variables and tools 25
3.0.8 Research instruments 27
3.0.9 Study procedure 28
3.0.10 Data management and analysis 28
3.0.11 Quality control 29
3.0.12 Data safety 29
3.0.13 Ethics 30
3.0.14 Utility and dissemination 30
CHAPTER FOUR: 31
4.0 RESULTS 31
CHAPTER FIVE 40
5.0 DISCUSSION 40
REFERENCES 44
APPENDICES 51
APPENDIX I: INFORMED CONSENT FORM 51
APPENDIX III: STUDY INSTRUMENTS 53
The MINI 53
PATIENT HEALTH QUESTIONNAIRE-9 56
SOCIODEMOGRAPHIC QUESTIONNAIRE 58

List of Tables
Table 1: Socio-demographic characteristics of the study participants……………………………………………………33
Table 2 Biological/ clinical characteristics of the study participants ………………………………………………………34
Table 3: Diagnosis of depressive illness among patients with TB……………………………………………………………35
Table 4: Social factors associated with depressive illnesss among participants with TB………………………..37
Table 5: Biological/ clinical characteristics associated with depressive illness among patients with TB………………………………………………………………………………………………………………………………………………………….38
Table 6: Multivariate logistic regression analysis for the factors associated with depressive illness among the patients with TB………………………………………………………………………………………………………………………………39

List of Figures
Figure 1: Conceptual frame work of factors associated with major depression in patients with tuberculosis………………………………………………………………………………………………………………………………17
Figure 2: Severity of depression among participants with depressive illness on the MINI…………………………………………………………………………………………………………………………………………36

DECLARATION
I hereby declare that all the work submitted is original except where otherwise stated. This work has not been presented to any other university for a degree nor has it been submitted anywhere for publication.

SIGNED………………………………………………DATE…………….
Dr. Racheal Alinaitwe MBChB (MAK)
(Student)

SUPERVISORS
SIGNED………………………………………………DATE…………….
Associate Professor Noeline Nakasujja MBChB, MMED, PhD

SIGNED………………………………………………DATE…………….
Dr. Paul Bangirana BSc, MSc, PhD

SIGNED………………………………………………DATE…………….
Dr. Caroline Birungi MBChB, MMED

DEDICATION
This work is dedicated to my husband; Mr. Horeb Rukiri, my son; Timothy Rukiri and my dear mother; Ms Harriet Nyamahunge who endured long periods of my absence during the preparation of this work.

ACKNOWLEDGEMENT
My foremost appreciation goes to the Almighty God who enabled this work to be compiled and successfully brought to completion for the award of the masters’ degree in Medicine.
I would sincerely like to acknowledge my supervisors; Assoc. Professor Noeline Nakasujja, Dr. Paul Bangirana and Dr. Caroline Birungi. This work has been a success because of their tireless efforts in advising, correcting and encouraging me every step of the work. I will forever be indebted to their generosity in sharing their knowledge and experience.
I also wish to acknowledge the Department of Psychiatry Makerere University and the staff of Global Health Uganda for the financial support they rendered me during my course hence making this work possible.
My appreciation further goes to the Head and Staff of the Tuberculosis unit in Mulago Hospital for their support in the proposal development and data collection process.
My heartfelt gratitude goes to my mentor Dr. Catherine Abbo for her constant encouragement and advice in my tough and gloomy days of this course and the carrying out of this research.
My classmates; Dr. Ritha Binduru (Sr), Dr. Hillary Kuteesa and Dr. Leticia Kyohangirwe thank you for the ideas and encouragement through the course.
I also appreciate my research assistants; Mr. Edson Atwine and Mr. Moses Muhangi and my biostatistician Miss Peace Aber.
My humble appreciation goes to the patients of tuberculosis who gave us their time to interact with us during the study. These patients suffer from a double burden of depression and tuberculosis and yet unrecognized by many who care for them.

OPERATIONAL DEFINITIONS
Depressive illness: this includes major depression and dysthymia according to the Mini Neuropsychiatric Interview (MINI).
Major depression: five (or more) of the following symptoms have been present during the same 2-week period and represent a change from previous functioning; at least one of the symptoms is either depressed mood or loss of interest or pleasure and other symptoms include; weight loss or weight gain, decrease or increase in appetite, insomnia or hypersomnia, psychomotor agitation or retardation, fatigue or loss of energy, feelings of worthlessness or excessive or inappropriate guilt, diminished ability to think or concentrate, suicidal behavior (APA 2013).
Dysthymia: at least 2 year history of depressive symptoms with no more than 2 months of no symptoms (APA 2013).
Tuberculosis treatment success rate: is the percentage of all new TB cases (or new and relapses) registered under a national tuberculosis control programme in a given year that successfully completed treatment with or without bacteriological evidence of success.
Tuberculosis case: a patient with tuberculosis diagnosed clinically, by laboratory or radiological investigation.

List of Abbreviations
3TC………………………………..Lamivudine
AIDS……………………………..Acquired Immunodeficiency Syndrome
ART………………………………Antiretroviral Therapy
COPD…………………………….Chronic Obstructive Pulmonary Disease
DALYs……………………………Disability Adjusted Life Years
DOT……………………………….Directly Observed Therapy
DSM……………………………….Diagnostic and Statistical Manual for Mental Disorders
EFV………………………………..Efavirenz
HIV…………………………………Human Immunodeficiency Virus
ICD…………………………………International Classification of Diseases
MDR-TB……………………………Multidrug Resistant Tuberculosis
MINI…………………………………Mini-International Neuropsychiatric Interview
NTLP………………………………..National Tuberculosis and Leprosy Programme.
PHQ…………………………………Patient Health Questionnaire
SCID-P…………………………….. Structured Clinical Interview for DSM-III-R Patients
SPSS………………………………..Statistical Package for Social Scientists
TB…………………………………..Tuberculosis
TDF…………………………………Tenofovir
YLDs………………………………..Years Lived with Disability
ZN…………………………………..Ziehl-Neelsen
ABSTRACT
Background: Depression is a major cause of the global disease burden and globally affects 350-400 million persons making it the largest contributor to years lived with disability. Among patients with chronic physical illnesses like tuberculosis (TB) especially in developing countries, depression can affect up to 25-33% of patients. More than 1.5 million tuberculosis cases occur in Sub Saharan Africa yearly. Uganda is one of the world’s high TB/ Human Immunodeficiency Virus (HIV)- burden countries with a 2015 prevalence of all TB cases in Uganda found at 202 per 100.000.The treatment success rate was 75 % for new and relapse cases and 73% for HIV-positive TB cases.
Objective: To determine the prevalence and factors associated with depressive illness in patients with tuberculosis in Mulago Hospital.
Methods: This was a cross sectional study involving participants aged 18 years and above diagnosed with tuberculosis attending the tuberculosis clinic in Mulago Hospital. Consecutive sampling was done until the sample size of 308 participants was achieved. Participants were administered the Socio-demographic, the Mini Neuropsychiatric Interview (MINI) to diagnose depressive illness and the Patient Health Questionnaire- 9 for rating the severity of depression and. Data was entered using Epi-Data. Descriptive, bivariate and multivariate analysis was done with SPSS.
Results: The prevalence of depressive illness was 23.8% (95% confidence interval 19.32 -28.89) as per the MINI. Prevalence of major depression was 23.8% and that of dysthymia 2.27% (95% C.I 1.11-4.81). Depressive illness was associated with low education level (p= 0.003), bein g in the intensive phase of TB treatment (OR=2.344, p=0.007) and family history of depressive illness (OR=5.422, p=0.001).
Conclusion: The prevalence of depressive illness in patients with TB is high and is associated with low education level, being in the intensive phase of TB treatment and family history of depressive illness. Depressive illnesses should be screened and managed among patients with TB.
CHAPTER ONE: INTRODUCTION
1.0 BACKGROUND
Depression is a major cause of the global disease burden, affecting an estimated 350-400 million persons worldwide. This makes depression the largest contributor to years lived with disability globally (Marcus et al 2012). Depressive disorders are the second leading cause of years lived with disability (YLDs) and Major Depressive Disorder accounts for 8.2% (5.9%–10.8%) of global YLDs in 2010 and dysthymia for 1.4% (0.9%–2.0%) (Murray et al 2013, WHO, 2001). Furthermore, in 2010, depression was estimated to cost at least US$ 800 billion in lost economic output and this sum is expected to more than double by 2030 (WHO, 2008). A research carried out by the World Health Organization (WHO) in 2012 in 17 countries revealed that one in 20 people reported depression (Marcus et al 2012) . At its worst, depression can lead to suicide and about 800,000 to 1 million people commit suicide annually (Marcus et al 2012). Major Depressive Disorder explained 16 million suicide Disability Adjusted Life Years (DALYs) (Murray et al 2013). Among patients with chronic physical illnesses like, Human Immuno-deficiency Virus/ Acquired Immuno-deficiency Syndrome (HIV/AIDS), hypertension, diabetes, cancers and tuberculosis (TB) the prevalence of depression is between 10-20% (Katon et al 2007). Depression and TB are an important public health concern that contributed 2.5 and 2% respectively of DALYs worldwide in 2010 (Murray et al 2013). Depression is three to six times more common in patients with tuberculosis compared with tuberculosis negative patients (Sweetland et al 2014).

Tuberculosis is a global health problem and in 2014, 9.6 million people developed TB and 3.3% were multi-drug resistant tuberculosis (MDR-TB) cases. More than 1.5 million TB cases occur in Sub Saharan Africa each year (Zumla et al 2015, Mweemba et al 2008). It is also worth noting that the prevalence of TB in Uganda in 2013 was 159 per 100.000 with 1400 new cases of MDR-TB and a treatment failure rate of 25% (Zumla et al 2015). The current prevalence of depression among patients receiving TB treatment has a wide range from 11.3% to 80.2% with a mean weighted prevalence of 48.9% (Sweetland et al 2014). This wide range could be explained by use of different methods in these studies with some using brief screening instruments and others using clinical diagnostic interviews. Evidence from cross sectional studies done in African hospitals indicate a very high prevalence of comorbid depression among patients with TB ranging between10 and 52% with the one study in Nigeria reporting it at 27.7% (Deribew et al 2010).
As with psychiatric illnesses, tuberculosis also carries social stigma the world over, a proportion of society also considers tuberculosis to be incurable hence the patients endure neglect and psychological trauma once diagnosed. It is highly probable that this trauma may predispose them to psychiatric disorders (Basu et al 2012). The relationship between depression and chronic physical illness is bidirectional (Katon, 2011). Patients with TB may develop depression due to chronic infection or related psychosocial stressors or as a result of effects of drug like isoniazid (Mitnick et al 2008, Mikkelsen et al 2004). The compromised immunity and reduced self-care associated with depression could also predispose an individual to contracting TB directly (Mikkelsen et al 2004). Depression also predisposes to risky behavior with resultant contraction of HIV which lowers immunity hence one easily gets tuberculosis as well (McFarlane et al 2014, Prince et al 2007). This comorbidity is associated with a range of adverse outcomes including functional impairment, increased medical costs, poor adherence to medication, community transmission, increased medical symptom and pill burden hence increased mortality (Katon, 2011). Individuals who are depressed with TB are less likely to seek care promptly and while in care are less likely to consistently and completely take their medication, yet poor adherence to anti-TB medications is recognized as a significant cause of treatment failure (Pachi et al 2013, Namukwaya et al 2011). Inconsistent and incomplete treatment of TB results in treatment which is significantly more expensive, takes about four times longer duration, and produces acute physical and psychiatric side effects which makes treatment adherence and completion a much bigger challenge (Aamir, 2010, Sweetland et al 2014). Adherence is especially important in MDR-TB patients as this is most times the last treatment option and failure to complete this treatment may lead to high rates of fatality in addition to ongoing transmission of highly resistant drug strains (Sweetland et al 2014). Therefore, early diagnosis and timely intervention of depression in patients with TB may improve adherence to anti-TB medication (Sweetland et al 2014,Aamir, 2010).
HIV has also been noted as a driving force behind the global burden of TB. In 2000, about 11.5 million people with HIV were co-infected with TB and 70% of these were in sub-Saharan Africa. (WHO, 2004). Both TB and HIV are chronic physical illnesses that have psychological effects on the sufferers due to the diseases themselves, medications and stigma associated with both conditions (Deribew et al 2010). In addition, people with depression are also noted to be at an increased risk of contracting HIV/AIDS (Prince et al 2007). Patients co-infected with TB and HIV are most times subject to potentially toxic treatments which may cause debilitation, stress and demotivation (Isaakidis et al 2013). Discrimination as a result of the dual diagnosis, dependence on family for support and financial problems also affect the patients’ mental wellbeing with resultant psychiatric disorders (Ownby et al 2010). In Uganda depression in HIV positive patients is estimated to range between to 47% to 60% (Singh et al 2008, Kaharuza et al 2006). Depression in HIV positive patient is estimated at 2-3 times than in the general population and this population also has more severe degrees and symptoms of depression (Nakasujja et al 2010, Akena et al 2010, Sherbourne et al 2000).

A study done in Uganda at Mulago Hospital on the prevalence of depressive symptoms among medically ill patients found that 25% of the patients had chest infections including pulmonary TB and 35% of the medically ill patients had depressive illness (Katorogo, 2006). This study however had only a sample size of 60 patients and the prevalence of major depression among patients with TB specifically was not reported. Furthermore there is limited data on the prevalence and factors associated with depressive illness in patients with TB in the Ugandan setting. Therefore this study will seek to determine the prevalence of depressive illness in patients with TB attending the TB clinic in Mulago Hospital and also find out the biological and psychosocial factors that may be associated with the two conditions.
1.1 PROBLEM STATEMENT

Uganda is one of the world’s 41 high TB/HIV- burden countries. These are countries with the highest estimated numbers of incident TB cases that account for 97% of the global TB burden. They also have the highest burden of TB in terms of per capita expenditure (WHO, 2015).
The prevalence of all TB cases in Uganda in 2015 was 202 per 100, 000 with the incidence of MDR-TB at 4.9 per 100,000 and of TB/HIV at 66 per 100,000. In 2015, Uganda had a treatment success rate of 75 % for new and relapse cases and 73% for HIV-positive TB cases (WHO, 2015). About 29.4% of people with TB are also infected with HIV and this comorbidity is known to be associated with psychiatric illness due the stigma, chronicity and psychological distress of the illnesses (Ali et al 2016).
Tuberculosis is associated with psychiatric morbidity, particularly depressive disorder, which has been recognized as a cause of poor compliance with resultant treatment failure and emergence of drug resistant TB strains (Issa et al 2009).
Despite the knowledge that the comorbidity of depression with TB is common, little effort has been made to the identification of depression among TB patients. Indeed there is little evidence of programs for TB care that have mental health as an option. A number of factors have been shown to be associated with depression among patients with TB and these include: severe symptoms of TB, type of TB, presence of HIV/AIDS, being single, divorced or widowed, age, being unemployed, long duration of illness and category of TB treatment (Kehbila et al 2016, Pachi et al 2013, Erhabor et al 2000).

Without addressing the burden of depressive illness among TB patients, adherence and prevention of disease transmission is bound to remain an enormous challenge in our setting. This study set out to determine the prevalence and associated factors of depressive illness in patients with TB attending the TB clinic in Mulago Hospital, hence provided important information for management and policy formation.
1.2 SIGNIFICANCE
Studies about the comorbidity of depressive illness and TB have been done in India, Nigeria, Cameroon, Ethiopia and Tanzania however a study done in Uganda focused mainly on depression in medical conditions generally. There is limited data on the association between depressive illness and TB infection.
This study contributes to our understanding of a growing public health problem, which is the double burden of TB and depression. The results of the study provide information on prevalence and factors associated with depression among patients with TB. This information therefore provides a basis for integration of psychiatric management in patients with tuberculosis and as such will improve the quality of care of these patients and possibly result in improved TB treatment outcomes.
1.3 RESEARCH QUESTION

1.3.1. What is the prevalence and factors associated with depressive illness in patients with tuberculosis in Mulago Hospital?

1.4 OBJECTIVES

1.4.1 GENERAL OBJECTIVE
To determine the prevalence and factors associated with depressive illness in patients with TB in Mulago Hospital.
1.4.2 SPECIFIC OBJECTIVES
1. To determine the prevalence of depressive illness in patients with TB in Mulago Hospital.
2. To determine the biological and psychosocial factors associated with depressive illness in patients with TB in Mulago Hospital.

1.5 CONCEPTUAL FRAMEWORK
Conceptual framework showing how different factors are associated with depressive illness in patients with TB. This conceptual framework is based on the bio-psychosocial model (Borrell-Carrió et al 2004). According to this model, disease is a result of an interaction between biological, psychological and social factors.

The above conceptual frame work highlights factors associated with depressive illness in patients with tuberculosis however this study majorly focused on the social factors like; age, sex, employment status, education level, marital status, biological/ clinical factors like; type of TB, category of treatment, phase of TB treatment, family history of depression, substance use, and HIV/AIDS. We did no assess any psychological factors.
CHAPTER TWO
2.0 LITERATURE REVIEW
2.0.1 Magnitude of depression among patients with TB
Depression is a common mental disorder characterized by a two week history of loss of interest in pleasurable activities, feelings of sadness, feelings of guilt or worthlessness, low sleep or appetite, low energy and poor concentration, insomnia or hypersomnia and suicidal or homicidal behaviors (Karthik et al 2016, APA, 2013, Marcus et al 2012 ). This psychiatric condition arises from an interplay of biological, psychological and social factors. Biological factors include medical illness abnormalities in the biogenic amines, other neurotransmitters and genetic predisposition. The psychosocial factors point to the fact that environmental stressors and adverse life events are a contributor to development of depressive illness (Katon, 2011). Research is showing an increase in the association between mental health problems especially depression and chronic physical illnesses like TB yet research in low and middle income countries in this area is still limited (Buberwa, 2013).
A prospective study conducted at a tertiary care hospital in South India from March 2014 to
June 2015 involving 147 patients registered in the Directly Observed Therapy (DOT) clinic revealed that 84% of patients who were currently on anti-TB drugs for at least one month duration had some degree of depression on the Patient Health Questionnaire-9 (PHQ-9). Moderate depression was the most common in this study occurring among the unmarried, patients experiencing side effects, persistent coughing and weight gain at the time of the study (Rahul 2015).
Another study carried out at a DOT Centre in Nigeria to assess depression comorbidity among patients with tuberculosis in a university teaching hospital found the prevalence of depression to be 27.7% of which 21.5% had mild depression and 6.2% had moderate depression. The clinical factor of persistent cough was significantly associated with depression on the PHQ-9 (p=0.04) among 65 patients in that study (Issa et al 2009).

A cross sectional study carried out in a district and regional hospital in Cameroon in 2015 on 265 patients with pulmonary tuberculosis using the PHQ-9 found a prevalence of depression of 61.1 % (95 % CI: 55.1–66.8), with a significant proportion (36.6 %) having mild depression and 24.5 % had moderate depression. This study included both in and out-patients (Kehbila et al 2016). In another cross sectional study carried out in South Ethiopia using the Hospital Anxiety and Depression Scale on 417 patients at a University Hospital and a Health Center, prevalence of depression was found at 43.4 % (Duko et al 2015).
In yet another study conducted in Tanzania on 390 participants using the same instrument, PHQ-9, the prevalence of depression in TB patients who were HIV negative was found to be 46.9% with 33.6 % having mild depression and 13.3% having moderate depression. This study also found that being in the intensive phase of treatment was associated with mild and moderate depression However this study excluded patients with HIV/AIDS (Buberwa, 2013).
2.0.2 Possible mechanisms for the association between TB and depression
The pathways for the association between TB and depression are complex and multidirectional. These pathways are both biological and psychosocial however the individual extent of contribution to the burden of comorbidity is currently unclear (Mitnick et al 2008). It has been suggested that patients with TB may develop depression due to chronic infection or related psychosocial stressors or as a result of effects of drug like isoniazid (Mitnick et al 2008, Mikkelsen et al 2004). Serious medical illness is a significant stressor that affects body image, self-esteem, capacity to maintain family and social relationships, and sense of loss one experiences (Mikkelsen et al 2004).

The compromised immunity and reduced self-care associated with depression could also predispose an individual to contracting TB (McFarlane et al 2014).The association between depression and the severity of TB symptoms could also be because depression increases the severity of symptoms of physical illness (APA,2013). It has also been noted that these two comorbidities may share risk factors like depressed immunity in presence of other chronic infections like diabetes and HIV/AIDS (Kiecolt-Glaser et al 2002).
Immunologically chronic infections lead to overproduction of inflammatory cytokines like interleukin 6, resulting in endocrine reactions that cause depressive symptoms (Kiecolt-Glaser et al 2002). Depression also enhances the production of inflammatory cytokines which down-regulate the patients’ immunity making them susceptible to TB and other chronic medical illnesses (Sweetland et al 2014, Kiecolt-Glaser et al 2002). Being a chronic pulmonary condition, TB is associated with hypoxia which can lead to depression (Mikkelsen et al 2004) . More so chronic diseases like TB are associated with weight loss, fatigue, psychological and social losses and these may trigger depressive reactions (Mikkelsen et al 2004). People who have chronic diseases and comorbid depression can benefit from treatments for depression, including use of antidepressants and psychotherapy (Doherty et al 2013, Rayner et al 2010, Simon et al 2005).
2.0.3 Factors associated with depression among patients with TB
Biological factors
The prevalence of depression is associated with biological factors like duration and category of TB treatment. A study done in Turkey showed the prevalence of depression of 19% for recently diagnosed TB patients, 21.6% for defaulted TB patients, 25.6% for patients with multidrug-resistant TB and 47.3% in chronic obstructive pulmonary disease (COPD) (Aydin et al 2001).
Family history of mental illness was also found to be independently associated with depressive illness in this population. Patients with family history are about 2.5 times more likely to be depressed than those without family history (Kehbila et al 2016). This could be due to the fact that depression has genetic predisposition in its etiology.

TB/HIV co-infection lowers the patients’ quality of life in social, economic and physical domains hence also contributing to these patients developing psychological distress and other psychiatric disorders like depression and poor physical health generally. The comorbidity of TB and HIV increases the odds of developing common mental disorders like depression by 1.7 compared to patients with HIV alone and by 2.5 as compared to those with tuberculosis alone (Kehbila et al, 2016, Deribew et al 2010).TB/HIV co-infected patients can be at higher risk of common mental disorders as a result of stigma and discrimination by the society (Duko et al 2015). Some disease processes like HIV and tuberculosis directly affect the brain and can cause brain damage leading to cognitive impairment and mood disorders like depressive illnesses. (Prince et al 2007).

Psychological factors
An increase in the number of symptoms reported by the patients, more serious perceived consequences of TB and patients having a feeling of less control over their illness also increases the prevalence of depression in this population. A Pakistan study among 108 patients with TB showed that raised depression and anxiety scores were associated with the above mentioned factors (Husain et al 2008). Patients with perceived stigma as a result of tuberculosis have low self image and are more likely to be socially withdrawn making them about 11 times more prone to depressive illness than their counter parts (Duko et al, 2015).Chronic diseases like tuberculosis create a psychological burden due to factors like the acute trauma of the diagnosis, difficulty of living with the disease, threat of decline in function and shortened life-span in addition to the complicated therapeutic regimens (Prince et al 2007).

Social factors
Socio-demographic factors like being a widow(er) or single, unemployment, being accompanied to hospital and a clinical factor (persistent cough) also increase the comorbidity of TB and depression as shown by a prevalence study in Nigeria amongst 105 TB patients ( Issa et al 2009, Aniebue et al 2007). Increasing age was reported as a significant factor in developing depressive illness. A study in south Ethiopia showed that for every 14 years increase in age, the risk of having depression increased by 19.0% (95%CI = 1.06,1.33) (Ambaw et al 2017).
Being female is also shown to be associated with a higher prevalence of depressive illness in patients with TB (38.5 % vs. 22.6 %). This has been attributed to the association between female hormonal factors and depression (Ambaw et al 2017, Kehbila et al 2016). Poor social support among patients with tuberculosis was independently associated with depressive illness in a study done in South Ethiopia (Ambaw et al 2017, Duko et al, 2015). Patients with TB voluntarily separate themselves from their family for fear and guilt of spreading infection to other members. The impact of stigma reported has led to divorce, cancellation of impending marriages, breakdown of family relationships and also isolation within the family. Stigma also negatively impacts the patient in accessing healthcare facilities in their neighborhood (Thomas et al 2016).
The level of education also has significant bearing on the comorbidity because through education one can have access to a wide range of health advancing resources, like higher incomes (better jobs), higher prestige, hence better (mental) health outcomes (Ambaw et al 2017,Berchick et al 2012).
2.0.4 Effect of depression on TB management
Mental disorders like depression can delay help-seeking, reduce the likelihood of detection and diagnosis of other health conditions like TB (Rukundo, 2016). This results in patients with depression to have diminished health related quality of life (Rapaport et al 2005). Mental disorders especially depression cause long term disability and dependency and this affects how these patients will adhere to their medications for tuberculosis treatment (Prince et al 2007).
A multicenter study carried out in South Africa, Zimbabwe, Zambia and Tanzania to determine psychological distress in 1502 patients using the Kessler- 10 revealed that patients who had been coughing for at least two weeks prior to presenting to the clinic (n = 1402) and are hence likely to have transmitted more disease experienced a delay in seeking care. These patients had higher psychological distress than those who had not had a cough for at least two weeks (n = 100) K-10 score of 22 (15–29) vs. 13(11.75-17.25); p = 0.0002) (Theron, G., et al 2015). In South Africa, another study done among 166 TB patients showed psychosocial factors that had a negative influence on adherence to the Directly Observed Therapy (DOT) program were feelings of helplessness (10.9%), depression (64.3%), and inadequate social support (Naidoo et al 2010).

HIV-infection and heavy alcohol usage were also associated with a delay in seeking care especially amongst patients with culture-confirmed TB. These patients were more likely to experience social marginalization and have side effects from their anti-TB medication which may further worsen non-adherence (Theron et al 2015). Many of these patients turn to alcohol as a way of coping with the psychological effects of tuberculosis like depression, stigma and anxiety only to end up with poor treatment outcomes (Thomas et al 2016).

A systematic review done in South Africa showed that adherence in patients with tuberculosis is also affected by family pressure, insufficient social support, a fear of disclosure which are all present in patients with a dual diagnosis of tuberculosis and depressive illnesses (Kastien-Hilka et al 2016). Non-adherence was related to poverty, HIV co-infection, stigmatization, an unsupportive social and work environment, and feelings of helplessness and hopelessness (Naidoo et al 2009).
Some drugs used in treatment of especially MDRTB like Cycloserine, Ethambutol, Isoniazid and Ethionamide are associated with psychiatric side effects like depression, psychosis and anxiety which adds on the psychological distress of these patients and further worsens their adherence to medication and hence treatment outcome (Thomas et al 2016).

CHAPTER THREE
3.0 METHODS
3.0 .1 Study design
This was a cross-sectional study.
3.0.2 Study setting and site
The study was carried out at the TB clinic also known as the National Tuberculosis and Leprosy Programme (NTLP) clinic at Mulago Hospital in Kampala. The NTLP clinic at Mulago Hospital serves as both the national referral center (approximately one-third of patients are referred) and the largest TB treatment clinic in Kampala, the capital city of Uganda. This is a predominantly urban area with 25% of all TB cases that are reported to the NTLP each year. The clinic has an average of 400 patients per quarter and it runs five working days. The clinic runs from 7:00am to about 3:00 pm. Monday, Wednesday and Thursday are for handling susceptible (not drug resistant) TB cases. Tuesday is for MDR-TB cases and Friday deals with TB/HIV comorbid patients. On average per day there are 30-40 susceptible cases, 30-40 MDR-TB cases and 40-50 TB/HIV cases on the respective clinic days.
3.0.3 Study participants
Target population; these were patients with TB in Uganda.
Accessible population; these were patients with TB attending the TB clinic during the study period.
Study sample population; these were patients with TB attending the TB clinic during the study period who met the eligibility criteria.
Study unit; this was an individual patient who fulfilled the eligibility criteria and consented to participate in the study.

3.0.4 Selection criteria
? Inclusion criteria
All patients aged 18 years and above with TB attending TB clinic and who had consented.
TB cases by clinical diagnosis, sputum analysis using Ziehl Neelsen (ZN) stain or Genexpert, abdominal ultrasound, x-ray and tissue biopsy.
We excluded two participants because we were unable to communicate effectively with them.
3.0.5 Sample size estimation
The sample size for objective one was calculated using the Leslie Kish formula (Leslie Kish, 1965) as follows:
n= {Z2 (p)(1-p)} /e2
Where; n=sample size.
Z is a factor to guarantee distribution that is 1.96
e is sampling error.
p is a prevalence from a previous study closest to this setting.
The closest to this setting was 27.7% (Issa et al., 2009) from a study conducted at an outpatient clinic at a teaching hospital in Nigeria on depression comorbidity among patients with tuberculosis.
N=308 participants.
The sample size for the second objective was calculated using the A-priori formula (Hulley et al 2013) for associated factors as follows;
N ? {Z? ? P(1-P)(1?q1?1?q2)} ? {Z? ? P1(1-P1)1?q1 ? P2(1-P2) 1?q2}2 ? (P1-P2)2
P ?p1q1? p2q2
Using a study by Kehbila et al 2016 done in Southwest region of Cameroon. Treatment status was used for the calculation.
P1 is proportion of new TB cases with depression; 55.4%
P2 is proportion of retreatment cases with depression; 36.1%
q1 is proportion of new cases in the study; 63.6%
q2 is proportion of retreatment cases in the study
Z? is a factor to guarantee distribution; 1.96
Z? is the power of the study; 0.84
N ? 223
Hence I used the sample size of 308 which was bigger than 223 to capture an appropriate sample for both study objectives.
3.0.6 Sampling method
Participants for the study were recruited using the consecutive sampling method where every accessible participant who met the selection criteria was included in the study. Data was collected from an average of 10 participants per day for the 5 days that the clinic runs and this was done over a period of 6 weeks.
3.0.7 Study variables and tools
? Dependent variable; depressive illness in patients with TB as assessed with the PHQ-9 and the Mini Neuropsychiatric Interview (MINI) 6.0.0.
? Independent variables; socio-demographic factors included; age, sex, employment status, education level, marital status, and biological/ clinical factors included; type of TB, category of treatment, phase of TB treatment, family history of depression, substance use and HIV/ AIDS.
3.0.8 Research instruments
Two research assistants (psychiatric nurses) were employed and they worked hand in hand with the principal investigator (PI) and the staff of the TB clinic.
1. A researcher-designed questionnaire was administered to record information on participants’ age, sex, educational level, marital status, and employment status. The following Tuberculosis disease characteristics were assessed; phase of TB treatment (e.g. intensive or continuation phase), TB treatment category (category I which included all non drug resistant TB cases regardless of previous treatment outcome and category II which included drug resistant TB cases), HIV status, Antiretroviral Therapy (ART) regimen and type of TB (pulmonary or extra pulmonary or both). Patients’ TB cards and medical files were also used to obtain some of this information. Family history of mental illness and substance use history in the past three months were also assessed.

2. Depressive morbidity was measured using the MINI which is a short structured diagnostic interview which was developed for the Diagnostic and Statistical Manual of Mental Disorders-IV (DSM-IV) and International Classification of Diseases (ICD) by psychiatrists and clinicians in the United States and Europe. Its average administration time is 15 minutes. It had a specificity of 0.88 and a sensitivity of 0.96 for major depressive episode when tested against the Structured Clinical Interview for DSM-III-R Patients (SCID-P) for a study done in Paris and Florida (Sheehan et al 1997). The MINI has been used in various studies in Uganda including a study by Abbo et al (2009) which assessed the prevalence and severity of mental illness handled by traditional healers in two districts in Uganda. It was also used by Kinyanda et al (2016) in the study on incidence and persistence of depression in HIV in Uganda.
3. The PHQ-9 was administered to all patients to rate the severity of their depressive symptoms.
The PHQ-9 was developed by Spitzer et al (1999). It is a questionnaire for primary care and was developed according to DSM-IV diagnostic criteria based on the 9 depressive symptoms (Bian et al 2011). It has been validated in the African setting for instance, when used in Nigeria on university students the internal consistency (Cronbach’s ? coefficient) of each item in the PHQ-9 was 0.85, and it had a good test-retest reliability over a one month interval (P 1 Week 26 8.44
Family history of depression
No 286 92.83
Yes 22 7.17
Used any drug of abuse/substance in the last 3 months
No 272 88.31
Alcohol 31 10.06
Alcohol, Marijuana 1 0.32
Alcohol, Cigarettes 2 0.65
Cigarettes 1 0.32
Tobacco 1 0.32

Out of the 178 participants who were HIV sero positive, 172 (96.6%) were on ART and 6 (3.4%) were on cotrimoxazole only. One hundred fifty (86.71%) were on Lamivudine/ Tenofovir/ Efavirenz (3TC/TDF/EFV) while the remainder were on other combinations.
4.2 Prevalence of depressive illness
The prevalence of depressive illness was 23.7% (95% confidence interval 19.32 -28.89) using the MINI. On the PHQ-9, the prevalence of depressive illness was 34.4% (95% confidence interval 29.29-39.93). Using the PHQ-9 we found 2 (2.7%) participants had severe depression, 17(5.5%) were moderately severely depressed 26(8.5%) were moderately depressed 61(19.9%) were mildly depressed while 202(65.6%) had minimal or no depression.

Out of the 73 participants who had current depressive disorder, 14 (19.18%) (95% confidence interval 11.56-30.12) had recurrent depressive episodes.
Table 3: Diagnosis of depressive illness among patients with TB
Variables Frequency
(n=308) Proportions
(%) 95% Confidence Interval
Major depression (Current)
No 235 76.22 (71.10-80.68)
Yes 73 23.78 (19.32-28.89)
Dysthymia Current
No 301 97.73 (95.18-98.89)
Yes 7 2.27 (1.11-4.81)

The severity of depression for participants who were diagnosed with the MINI was rated on the PHQ-9 and found as shown in figure 1 below.

Figure 2: Severity of depression among the participants who had depressive illness on the MINI
4.3 Factors associated with the depressive illnesss among participants with TB.
Among the social factors; sex, age, marital status, employment and education level were independently associated with the major depressive disorder as shown in table 4 below.

Table 4: Social factors associated with depressive illnesss among participants with TB
Variables Normal
(n=235) Depressed
(n=73) OR (95%CI) p-value
Sex
Male 152(64.96) 36(49.32) 1
Female 83(35.04) 37(50.68) 1.91(1.11-3.24) 0.017
Age
;50 15(6.41) 13(17.81) 1
30-50 147(62.82) 36(49.32) 0.28(0.12-0.64) 0.003
1 Week 19(8.09) 7(9.59) 1.16(0.49-2.91) 0.745
Family history of depression
No 224(95.30) 62(84.93) 1
Yes 11(4.70) 11(15.07) 3.60(1.49-8.68) 0.004
Drug abuse/substance in the last 3 months
No 209(88.89) 63(86.30) 1
Yes 26(11.11) 10(13.70) 1.26(0.58-2.77) 0.549

?
4.4 Multivariate analaysis for the factors associated with depressive illness among patients with TB
All factors which had a p-value less than 0.2 were considered for the multivariate analysis.
Level of education, phase of TB treatment and Family history of Depression were significantly associated with depressive illness among patients with TB.
Table 6: multivariate logistic regression analysis for the factors associated with depressive illness among the patients with TB
Variables Adjusted Odds Ratio 95%CI p-value
Sex
Male 1
Female 1.692 (0.93 -3.09) 0.083
Age
;50 1
30-50 0.527 (0.26-1.09) 0.063
20 Major depression, severe

SOCIODEMOGRAPHIC QUESTIONNAIRE
1. Study number………………
2. Sex a) Male b) Female
3. Age in years………………..
4. Address, village and district…………………………..
5. Tribe ………………..
6. Education level; i) No formal education ii) P1-P7 iii) S1-S6 iv) Technical v) University
7. Marital status; i) Married/ Co-habiting ii) Never married iii) Widowed iv) Separated/ divorced
8. Employment status; i) employed (specify)………………………… ii) retired iii) peasant farmer iv) housewife v) student vi) unemployed viii)other (specify)………..
9. What is your HIV status; a) Negative b) Positive c) unknown
10. If POSITIVE, are you on ARVs; a) Yes b) No
11. If YES, which combination and for how long……………..………………….
12. Confirmation of TB; a) x-ray b) Z N test c) Genexpertd)abdominal ultrasound e)tissue biopsy f) clinical
13. Type of TB; a) Pulmonary b) Extra-pulmonaryc) Both
14. Category of TB treatment; a)Category I b) Category II c) Category III d) Category IV
15. Which phase of TB treatment is the patient? a) intensive phase b) continuation phase
16. Anti-TB medication currently used……………………………………………….
17. For how long have you been on these Anti-TB medications? …………………….
18. Have you ever missed your Anti-TB medication; a) No b) Yes
19. If YES, for how long……………………………………………………………
20. Any one in your family who has ever suffered from depression? a) No b) Yes
21. Have you used any drug of abuse/substance in the last 3 months? a) No b) Yes
22. If YES, specify…………………………………