SCHOOL OF PUBLIC HEALTH, COLLEGE OF HEALTH SCIENCES, UNIVERSITY OF GHANA FACTORS ASSOCIATED WITH DEFAULTS FROM TREATMENT AMONG TUBERCULOSIS PATIENTS IN WESTERN SIERRA LEONE BY JONATHAN JOSEPH GREENE 10430962 THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON JULY, 2014 IN PARTIAL FUFILMENT OF THE REQUIREMENT FOR University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh i DEDICATION This work is dedicated to the Almighty God for giving me the wisdom to fulfill this dream. This work is also dedicated to my wife, Juliana Esther Greene, and children for their immeasurable support for coping with the stress associated with my absence from home. Dr Yvonne Harding who encouraged me to enroll into this study and has been a source of support in diverse ways to see the fulfillment of this work. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh ii DECLARATION This work is the result of an independent investigation under the supervision of Dr. Samuel Sackey. Where my work is indebted to the works of others, I have fully acknowledged. I declare, therefore that this dissertation has not been presented elsewhere, either in part or in whole for another degree. Signed by Resident ………………………………………… Jonathan Joseph Greene Supervisor …………………………………………………… Dr. Samuel Sackey University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh iii ACKNOWLEDGEMENT I appreciate the enormous support and dedication to my supervisor Dr. Samuel Sackey, your timely interventions are so much appreciated. To Dr. Yvonne Harding and staff of the German Leprosy and Tuberculosis Relief Association, Dr. Alie Wurie and staff of the National Leprosy and Tuberculosis control programme, staff of the Connaught Chest Clinic, Dr. Arnold Okoni-Williams for helping me with the analysis of the data, friends in research, you have been of great help, this work would not have been complete without your support. Dr. Brima Kargbo and staff of the HIV/AIDS, Secretariat. Finally, sincerely thank all respondents who participated in this study. Above all, I thank the Almighty God for bringing me this far. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh iv ABSTRACT Background: Successful treatment of tuberculosis (TB) involves taking anti-tuberculosis drugs for at least six months. Poor adherence to treatment means patients remain infectious for a longer period, and are more likely to relapse or succumb to tuberculosis and could result in treatment failure as well as foster emergence of drug resistant tuberculosis. Sierra Leone is among countries with high tuberculosis burden globally. Objective: To identify factors associated with default from treatment among tuberculosis patients in Western Sierra Leone. Method: A case-control study was conducted between September 2012 to August 2013 in Western, Sierra Leone. The study population consisted of tuberculosis patients on treatment who default (non-compliant) and those who have completed treatment (compliant) at the Connaught Chest Clinic. A sample size of 240, 120 for compliant and 120 for non compliant. Respondents completed structured questionnaire for socio-demographic characteristics, patient related factors, socio-economic variables, healthcare system related factors, default factors, stigma and discrimination and disease and medicine related factors. Factors associated with default were analyzed using logistic regression, bivariatr and multivariate analysis. Results: Among the 120 defaulters interviewed, 34.2% (41) attributed their default to inadequate supply of medicines, 30.8% (37) to feeling better, 11.7% (14) not feeling better, 8.3% (10) lack of family support, 6.7% (8) no reason, 5% (6) stigma and discrimination and 3.3% (4) side effects . On multivariate analysis, food availability (OR=6.93,p=0.01) , patient waiting time (OR =0.09,p<0.01) , availability of medicine (OR=0.006,p<0.01),treatment University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh v should be taken until six months(OR=0.007,p<0.01),leaving with friends(OR=5.85,p<0.01) were independently associated with default. Conclusion: Non-availability of medicines, non-availability of food, long waiting times at the clinic, perception that TB treatment should be taken until six months, type of person living with the patient ,duration of stay in the current abode, disease classification ,HIV status and patients taking other medicines besides TB treatment had significant association with default from treatment among TB patients in Western Sierra Leone. Key words: Tuberculosis, Compliance, Defaulter, Western Sierra Leone University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh vi Contents DEDICATION...................................................................................................................................................... i DECLARATION ................................................................................................................................................. ii ACKNOWLEDGEMENT .................................................................................................................................. iii ABSTRACT........................................................................................................................................................ iv LIST OF TABLES.............................................................................................................................................. ix LIST OF FIGURES ............................................................................................................................................. x CHAPTER 1 ........................................................................................................................................................ 1 1.0 INTRODUCTION ............................................................................................................................... 1 1.1 PROBLEM STATEMENT.................................................................................................................. 2 1.2 SIGNIFICANCE OF THE STUDY..................................................................................................... 3 1.3 OBJECTIVES...................................................................................................................................... 3 CHAPTER 2 ........................................................................................................................................................ 5 2.0 INTRODUCTION ............................................................................................................................... 5 2.1 TUBERCULOSIS AND RELATED CONDITIONS.......................................................................... 5 2.2 HOST-RELATED FACTORS FOR DEFAULT AMONG TB PATIENT......................................... 6 2.3 ENVIRONMENTAL RELATED FACTORS FOR DEFAULT AMONG TB PATIENTS............. 11 2.4 STRATEGIES USED TO CONTROL TUBERCULOSIS ............................................................... 15 CHAPTER 3 ...................................................................................................................................................... 20 3.0 STUDY SETTING............................................................................................................................. 20 3.1 STUDY DESIGN............................................................................................................................... 20 3.2 STUDY POPULATION .................................................................................................................... 20 3.3 SAMPLE SIZE .................................................................................................................................. 21 3.4 SAMPLING PROCEDURE .............................................................................................................. 22 3.5 STUDY PROCEDURE ..................................................................................................................... 22 3.6 ETHICAL APPROVAL .................................................................................................................... 23 3.7 DATA ANALYSIS............................................................................................................................ 24 3.8 LIMITATION OF THE STUDY............................................................................................................. 24 CHAPTER 4 ...................................................................................................................................................... 25 4.1 DEMOGRAPHIC FACTORS ........................................................................................................... 25 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh vii 4.1.1 RESPONDENTS’ GENDER (N : 240) ..................................................................................... 25 4.1.2 AGE OF RESPONDENTS (N=240) ......................................................................................... 26 4.1.3 RESPONDENTS’ RELIGION (N = 240) ................................................................................. 26 4.1.4 MARITAL STATUS (N-240) ................................................................................................... 27 4.1.5 EDUCATIONAL LEVEL ATTAINED (N = 240) ................................................................... 28 4.2 PATIENT-RELATED FACTORS .................................................................................................... 29 4.2.1 CIGARETTES SMOKING (N=240)......................................................................................... 29 4.2.2 DRINKING ALCOHOL (N =240) ............................................................................................ 30 4.2.3 AVAILABILITY OF TREATMENT SUPPORTER (N = 240)................................................ 30 4.3 SOCIO-ECONOMIC FACTORS...................................................................................................... 31 4.3.1 EMPLOYMENT STATUS (N=240) ......................................................................................... 31 4.3.2 INCOME LEVEL OF RESPONDENTS................................................................................... 32 4.3.3 FOOD AVAILABILITY AMONG RESPONDENTS (N=240) ............................................... 33 4.4 HEALTH CARE SYSTEM - RELATED FACTORS....................................................................... 34 4.4.1 CONVENIENT TB CLINIC OPENING TIMES (N=240) ....................................................... 34 4.4.2 WAITING TIME AT CLINIC (N=240).................................................................................... 35 4.4.3 DISTANCE TRAVELLED (N = 240)....................................................................................... 35 4.4.4 COST OF TRANSPORT (N = 240) .......................................................................................... 36 4.4.5 DOT STATUS (N = 240)........................................................................................................... 36 4.4.6 HEALTH WORKERS ATTITUDES TOWARDS PATIENTS (N= 240)................................ 37 4.4.7 MEDICINE AVAILABILITY (N = 240).................................................................................. 37 4.5 KNOWLEDGE OF TUBERCULOSIS AND TREATMENT .......................................................... 39 4.5.1 KNOWLEDGE OF SYMPTONS (N=240)............................................................................... 39 4.5.2 PERIOD OF TREATMENT FOR TB (N = 240) ...................................................................... 39 4.6 DEFAULT FACTORS ...................................................................................................................... 40 4.6.1 WHO RESPONDENTS LIVES WITH (N = 240) ................................................................... 40 4.6.2 HOW MANY OTHER PEOPLE DO YOU LIVE WITH (N = 240) ....................................... 41 4.6.3 SIZE OF RESPONDENTS DWELLING PLACE (N= 240)..................................................... 42 4.6.4 LENGTH OF STAY IN THE CURRENT PLACE (N = 240) ................................................. 42 4.7 STIGMA AND DISCRIMINATION ................................................................................................ 44 4.7.1 DISCLOSED TB STATUS (N =240)........................................................................................ 44 4.7.2 REASONS FOR NON-DISCLOSURE (N = 19) ...................................................................... 44 4.8 DISEASE AND MEDICINE RELATED FACTORS....................................................................... 44 4.8.1 EXPERIENCED SIDE AFFECTS TO TB MEDICINES (N = 240)......................................... 44 4.8.2 TYPE OF SIDE EFFECTS RESPONDENTS EXPERIENCED (N = 157).............................. 45 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh viii 4.8.3 BEFORE TIME IT TOOK YOU FELT BETTER? (N = 240).................................................. 46 4.8.4 RESPONDENT’S REASONS FOR NOT COMPLETING TREATMENT (N = 120) ............ 46 4.8.5 TB DISEASE CLASSIFICATION (N = 240) ........................................................................... 47 4.8.6 HIV STATUS (N = 240)............................................................................................................ 47 4.8.7 THE TAKING OF OTHER MEDICINCES BY RESPONDENTS (N = 240) ......................... 48 4.8.8 WHAT OTHER MEDICINES WERE TAKEN? (N = 87) ....................................................... 48 4.8.9 PATIENTS’ OPINION ON WHAT COULD MAKE TB PATIENTS COMPLETE TREATMENT ........................................................................................................................................... 48 CHAPTER 5 ...................................................................................................................................................... 52 5.1 INTRODUCTION ............................................................................................................................. 52 5.2.1. SOCIO-DEMOGRAPHIC CHARACTERISTICS.................................................................... 53 5.2.2 PATIENT RELATED FACTORS............................................................................................. 54 5.2.3. Socio-Economic Status .............................................................................................................. 55 5.2.4. Treatment literacy related-factors .............................................................................................. 55 5.2.5. Healthcare System related factors.............................................................................................. 57 5.2.6. Default Factors........................................................................................................................... 58 5.2.7 Stigma and Discrimination................................................................................................................ 59 5.2.8. Disease and Medicine related .................................................................................................... 59 CHAPTER 6 ...................................................................................................................................................... 63 6.1 Conclusion ............................................................................................................................................... 63 6.2 Recommendations.................................................................................................................................... 63 6.2.1 Recommendations for further research studies to be conducted by the National TB Programme... 63 6.2.2 Recommendations to decrease non-compliance to tuberculosis treatment ...................................... 64 REFERNCES..................................................................................................................................................... 65 ANNEX 1: QUESTIONNAIRE ........................................................................................................................ 73 ANNEX2: PATIENT CONSENT FORM ......................................................................................................... 81 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh ix LIST OF TABLES Table 4. 1 Demographic factors for case and control ........................................................................................ 29 Table 4. 2 Patient related factor for case and control......................................................................................... 31 Table 4. 3 Socio- economic factors for case and control respondents ............................................................... 34 Table 4. 4 Healthcare system related factors for case and control respondents................................................. 38 Table 4. 5 Knowledge and treatment of TB for case and control respondents ................................................. 40 Table 4. 6 Default factors in TB treatment for case and control respondents.................................................... 43 Table 4. 7 Summary of the distribution of side effects among the compliant and non- compliant group ......... 45 Table 4. 8 Reasons given for non-compliance, by cases (defaulters) ................................................................ 47 Table 4. 9 Patients opinion with regards to compliance .................................................................................... 49 Table 4.10 Multivariate analysis of factors associated with default …………………………………………..50 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh x LIST OF FIGURES Figure 4. 1 Sex distribution of cases and controls of TB treatment compliance .................................25 Figure 4. 2 Age distribution of case and control of TB treatment completion ....................................26 Figure 4. 3 Religion distribution of case and control of TB treatment completion .............................27 Figure 4. 4 Marital status of case and control to TB treatment completion.........................................28 Figure 4. 5 Educational levels attained case and control of TB treatment completion .......................28 Figure 4. 6 Employment status of case and control of TB treatment completion ...............................32 Figure 4. 7 Distribution of income level s among case and control of TB treatment completion .......33 Figure 4. 8 Respondents' preference for TB clinic times.....................................................................35 Figure 4. 9 A distribution graph of the person case and control live with...........................................41 Figure 4. 10 A distribution graph of the number of other persons case and control live with ............42 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh xi LIST OF ABBREVIATION AIDS – Acquired Immunodeficiency Syndrome ART – Anti-Retroviral Treatment DOTS – Directly Observed Treatment Short-course DR-TB – Drug- Resistant Tuberculosis HIV – Human Immunodeficiency Virus MDR-TB – Multi-Drug Resistant Tuberculosis NTCP – National Tuberculosis Control Programme NLTCP – National Leprosy and Tuberculosis Control Programme PTB SM+ - Pulmonary Tuberculosis Smear Positive PTB SM- - Pulmonary Tuberculosis Smear Negative TB – Tuberculosis TBCTA – Tuberculosis Coalition for Technical Assistance WHA – World Health Assembly WHO – World Health Organisation University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 1 CHAPTER 1 1.0 INTRODUCTION Tuberculosis is an infectious and transmissible disease caused by mycobacterium tuberculosis and occasionally by mycobacterium africanum and mycobacterium bovis, these are the main pathogenic species within the mycobacterium tuberculosis complex (Madigan et al). Tuberculosis affects the lungs in more than 80% of cases; these are called pulmonary tuberculosis (WHO, 2009). Tuberculosis spread from the primary lesions to other parts of the body via the blood stream and by lymphatic and bronchial systems, in this way affects any other organ causing extra- pulmonary tuberculosis. Transmission of the bacilli from person to person occurs exclusively by air. The primary source of infection is a person with pulmonary tuberculosis who coughs and spreads the infectious droplets. In this 21st century, the world remains challenged by infectious diseases and tuberculosis in particular. Of the countries that have 80% of the world’s tuberculosis burden 17 are classified as low income countries (WHO, 2008). Annually, 2.3 million people still die from the disease, while increased deaths due to drug resistant tuberculosis, as well as co-infection of tuberculosis and HIV produce even higher threats to the world. Tuberculosis remains a global emergency causing high morbidity and mortality in Sub-Saharan Africa (WHO, 2008:3). In Sierra Leone, about 7,000 cases are detected annually (NLTCP, 2011). Successful treatment of tuberculosis involves taking anti-tuberculosis drugs for at least six months. Sierra Leone subscribes to the internationally accepted World Health Organization (WHO) strategy for tuberculosis control. In addition, the country has adopted the WHO recommended tuberculosis treatment regimen. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 2 In the first two months of treatment (intensive phase), a combination dose of rifampicin, isoniazid, pyrazinamide and ethanbutol are used daily, followed by four months of rifampicin and isoniazid ( WHO:2008) Some patients fail to adhere to treatment and eventually default before completing the treatment regimen. Patients whose treatment are interrupted for two consecutive months or more, as defined by WHO are reported as default, at the end of the treatment period. Poor adherence to treatment means that patients remain infectious for longer and are more likely to relapse or succumb to tuberculosis. Erratic or selective compliance to treatment and default could result in treatment failure, foster emergence of drug resistant tuberculosis and may increase the cost of treatment. According to Muture et al (2011) several risk factors associated with default from TB treatment were discussed. These risk factors are; social and economic factors, patient-related factors, health care and system related factors, and condition related factors. In a study done in India by Vijay et al (2010) on risk factors associated with default from TB treatment, they concluded that some of the risk factors are personal and socio-demographic risk factors and treatment related risk factors. The global default rate for new smear positive was 5%, for Africa the default rate was 8% and Sierra Leone have a default rate of 6% (WHO, 2009). 1.1 PROBLEM STATEMENT In Sierra Leone, treatment default is a serious problem in tuberculosis control. In 2011, default rate increases from 7.2% to 10.8% in 2012. The rise in default is of public health University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 3 concern, as this may lead to persistence of infectious source, increased mortality, increase relapse rate, and facilitate the development of resistant strains. Therefore, there is need to identify factors associated with tuberculosis treatment default in Western, Sierra Leone. 1.2 SIGNIFICANCE OF THE STUDY Identifying the factors associated with treatment default in Western Sierra Leone, will give insights into the reasons behind high defaulter and subsequently low treatment success rates. This will not only contribute to the existing body of knowledge on factors associated treatment default in general, but will also bring new knowledge specifically for Western Sierra Leone. Once these factors are identified, then targeted strategies to address them can be formulated. The study will benefit the tuberculosis patients (current and future) as findings may be used to formulate strategies to improve the quality of care. The Ministry of Health and Sanitation in Sierra Leone will also benefit as the findings of the study can be used as a basis for further generalizable studies. Recommendations could be made to the National Tuberculosis Control Programme on how tuberculosis treatment default can be improved in Sierra Leone. 1.3 OBJECTIVES The general objective for this study will be: - To identify factors associated with default from treatment among tuberculosis patients in Western Sierra Leone. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 4 The specific objectives for this study are: - To determine the demographic factors relating to treatment status among tuberculosis patients in Western Sierra Leone. - To determine host-related factors associated for default among tuberculosis patients in Western Sierra Leone. - To determine environmental-related factors associated for default among tuberculosis patients in Western Sierra Leone. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 5 CHAPTER 2 LITERATURE REVIEW 2.0 INTRODUCTION In keeping with the aim of this study, namely to identify factors associated with poor compliances TB treatment in the Western District of Sierra Leone and to make recommendations on how TB treatment compliance can be improved, this chapter reviews relevant literature from previous studies on factors associated with or contributing to TB treatment compliance or non-compliance. The scope of this literature review, therefore, is to synthesize evidence from textbooks published research, scientific reports and other credible sources of scientific work done globally, mainly on TB treatment compliance issues. 2.1 TUBERCULOSIS AND RELATED CONDITIONS According to WHO (1999:14), overcrowding, living in poorly ventilated dwellings and being in close contact with an infected individual are some of the risk factors to acquiring TB infection. When one is infected, poor nutrition and immune suppression predispose an individual to developing active TB disease (WHO 1999:15) In Sierra Leone, about 7% of all TB patients are HIV positive (NLTCP: 2011). Caminero (2003:69) recommends that the diagnosis of PTB should be based on the clinical presentation with confirmation by sputum smear microscopy and/or culture. The author further recommends that diagnosis of TB should always include performance of microbiological investigations for confirmation. Making an accurate diagnosis of TB is important as complying with TB medication is a heavy burden for the patient and the family in terms of cost, potential side effects, and stigma. A wrong diagnosis may delay appropriate treatment University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 6 for the patient and be fatal. Therefore, greater efforts should be made to ensure early and correct diagnosis, initiation of correct treatment and compliance to treatment to increase the chances of cure or treatment success (Bonilla et al 2008:2957). The emergence of drug resistant (DR) TB presents significant challenges to global TB control with an increase in incident cases reaching 489,000 MDR TB cases by 2006, representing a 65% increases since 2000 (WHO 2008c:13). Research has shown that failure to adhere to principles of TB control causes the development of almost all the DR TB, and poor or noncompliance to TB treatment is the main predisposing factor for an individual to develop DR-TB. This leads to treatment failure and subsequently may lead to death and further spread of DR-TB (Bonilla et al 2008:2957; Lalloo 2010:255). Given that current treatment success for MDR TB is about 60%, focus should be on preventing emergence of new DR-TB cases by ensuring proper management of drug susceptible cases (Orenstein et al 2009:153). According Gandhi et al (2006:1575) and Lalloo (2010:258), being infected with HIV is clearly a risk factor for developing DR TB, most likely due to intermittent TB treatment and poor absorption of anti-TB drugs. 2.2 HOST-RELATED FACTORS FOR DEFAULT AMONG TB PATIENT Previous studies relating to the factors affecting TB treatment compliance have identified the following host-related factors: patient literacy or educational level, alcohol and substance abuse, knowledge of TB and treatment (treatment literacy), feeling better, socio-economic factors, patients attitude towards treatment and co-morbidities (Kudakwashe:2010). Several studies have tried to look at the relationship of the patient’s education level to their health status, seen as important to gain a better understanding of the causes associated with adverse health outcomes, identifying patients at risk of such adverse outcomes and University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 7 subsequently developing appropriate interventions (De Walt et al 2004:1236). A study carried out in Thailand aimed at determining the patient factors predicting successful treatment. Out of 1,241 patients studied, 81% with higher educational levels and knowledge of tuberculosis were successfully treated, the argument being that these factors are associated with better compliance to TB treatment and subsequently treatment success (Okanurak et al 2008:1162). Several other studies have demonstrated educational levels of TB patients as significant predictors of treatment compliance (Balasubramanian et al 2004:352; Date et al 2005:680; Johansson et al 1999:868; Mishra et al 205:1134). Meanwhile, a Malaysian study demonstrated that, among other factors, non-compliance was associated with completed secondary education (O’Boyle et al 2002:307). Conversely, a study in Ndola (Zambia), found that age, marital status, and educational levels were not significantly associated with compliance (Kaona et al 2004:68). Studies have also showed that there is some relationship between default and alcohol intake and substance abuse. Alcohol and substance abuse have often been cited as reasons for poor compliance to medication in general. The altered behaviour under the influence of alcohol and other substances is believed to be the reason for such observations. When one is under the influence of alcohol one is likely to forget to take the medicines, and if even if not the chances of developing side effects that may subsequently lead to poor compliance are high (Sansone et al 2008:43). According to a study in Tomsk (Russia), the authors identified substance abuse as a barrier to TB treatment and care as it leads to non-compliance, default and acquisition of MDR TB (Gelmanova et al 2007:649), while in Uzbekistan, alcoholism and homelessness were associated with TB treatment default (Hasker et al 2008:97). Sansone et al (2008:45) assert that physicians and other healthcare workers treating patients need to be wary of the potential impact on treatment compliance of alcohol and substance use. In other words, they University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 8 should always explore alcohol and substance use in all their patients and any indications of such behaviour would enable the physician to focus their treatment literacy on such patients. Fry et al (2007:1027), insist that DOTS programmes are more likely to achieve better TB control outcomes if they include interventions aimed at improving diagnosis of alcohol and substance abuse and treating it concurrently with TB. Treatment literacy has also shown to be a factor associated with default among TB patients by some research done previously. According to Smart (2010:1), knowledge and attitudes about TB and its treatment vary widely due to different cultural, religious or traditional beliefs, and access to education and information about the disease. Smart (2010:1), further states that, patients’ lack of knowledge of TB symptoms or failure to recognize them, results in delays in seeking health care. Denial may be high due to stigmatization amongst misinformed communities. The above become barriers to early diagnosis and treatment, resulting in increased risk of transmitting TB to other close contact and the general community, as well as poor health outcomes for people with the disease (Afari-Twunamasi 2005:25). According to DeWalt et al (2004:1238), lack of treatment literacy is associated with poor health outcomes and conversely treatment literacy improves health outcomes and compliance. In Botswana, research found that compliance to treatment was related to the availability of information, material and emotional support from family members (Kgatlwan et al 2005:6) Several studies in India, Indonesia, Russia, (Woith et al 2009:1), Kenya, Tanzania, (Wandwalo et al 2000:1041) and South Africa (Afari Twunamasi 2005:23), have shown that knowledge of TB is generally low in many settings, even among healthcare workers. The study in India (Singla et al 1998:1005) surveyed 200 nurses and found that only 40% of TB nurses and 10% of general hospital nurses had a satisfactory knowledge of TB, and only 56% University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 9 of general nurses knew that TB was caused by mycobacterium tuberculosis. About 36% thought TB was caused by a virus, while in the Indonesian study only 40% of nurses knew its cause (Wahyuni et al 2007:135). In Kenya, Ayaya et al (2003:83) showed that most private medical practitioners were unaware of the correct methods of diagnosing TB and most used treatment regimens not recommended by the National Leprosy and TB Control Programme (NLTCP). Since most people with TB and their families get to know about TB from their healthcare providers, such poor knowledge of it among the providers will translate to lower knowledge among the patients themselves (Afari Twunamasi 2005:25). Often, when patients commence treatment they will be very sick and may be inactive. However, as the treatment progresses and their condition improves, and symptoms start to regress, the improvement in itself may become a barrier to continuation of treatment. The patient might not see the need to continue with treatment when they are feeling better or well (Williams et al 2008:73135). In a Nepal cross sectional study of 130 compliant and 25 non- compliant TB patients, 48% of the latter were more likely to think that they could stop TB treatment once they were free of symptoms and feeling well, because they thought they were cured (Bam et al 2005:51). Studies in Malaysia and Zambia showed that non-compliance was associated with being free of Symptoms (O’Boyle et al 2002:307; Kaona et al 2004:68). Patient’s defaulting behaviour occurs when the TB symptoms disappear and they feel well, usually after a few months of TB treatment (Pushpananthan et al 2000:20). Therefore, according to Bam et al (2005:56), the authors conclude that compliance to TB treatment could be improved by providing more information about TB to the patients. Compliance to TB treatment can, therefore, be improved by promoting TB treatment literacy among those with the disease, their families and communities through empowering the healthcare provider with knowledge of TB (Ndimande 2009:239). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 10 Socio-economic factors is also associated with default from treatment among TB patients, these factors will cover employment status socio-economic status and cost of transport. Differences exist as to whether employment and socio-economic status are contributory factors to patient TB treatment compliance (Pandit et al 2006:242). For some researchers, being employed may be associated with better socio-economic status, which enables one to afford cost of transport and healthcare fees, increasing the chances of treatment compliance (Okanurak et al 2008:1160; Tissera 2003:7; Hasker et al 2008:97). However, a study in India did not find socio-economic status to be significantly associated with TB treatment compliance (Pandit et al 2006:242). This study is located in the capital city of Sierra Leone and therefore it will be important to find out if socio-economic status has or has not any effect on TB treatment compliance. Studies in Nepal (Bam et al 2006:275), Uzbekhistan (Hasker et al 2008:97), Malaysia (O’Boyle et al 2002:307), Swaziland (Pushpananthan et al 2000:218), and Zambia (Needham et al 1998:811) indicated that cost of transport accounts for non-compliance to TB treatment, especially once the patient feels better. In the Malaysian study, cost and time of travelling to the treatment centre were major contributor factors associated with compliance to treatment, as non-compliant patients paid significantly more for transport than those compliant (O’Boyle et al 2002:310). A prospective cohort study in Southern Ethiopia to determine factors predicting treatment adherence among smear positive pulmonary tuberculosis patients found that among 404 TB patients on treatment, 20% defaulted treatment. In addition, 91% of all treatment interruptions occurred in the continuation phase, when the patient felt better and had higher cost of transport to a treatment facility (Shargie et al 2007:285). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 11 Self-efficacy, which is the belief that one can perform at a certain level to achieve certain goals, can determine a patient’s attitude towards treatment (Ormro 2006:367). If contextualized in the treatment for TB, the self-efficacy model could be a method for TB control focusing on motivating the patient to take treatment until completion. Healthcare providers, treatment supporters or family members, other patients currently taking treatment or those who took treatment previously and completed it, would act as motivators through verbal persuasion. Patients need to believe that if they comply with TB treatment they can achieve the desired outcome, which is to be cured (outcome expectations). Co-morbidities are illnesses occurring together, usually with one of the medical conditions or illness leading to the occurrence of the other. Psychiatric illnesses and HIV infection, as they are often homeless or have unstable housing conditions and lack food security, but they also frequently fail to comply with treatment for the same reasons (Fullilove et al 1993:324 31). A study in India however seemed to contradict the above findings as psychiatric disorders were not found to affect TB treatment compliance, whereas smoking tobacco did (Manoharam et al 2001:77). While there is no doubt that HIV predisposes an individual to developing TB, there is no consensus on whether HIV is associated with poor TB treatment compliance (WHO 2002:2). However, possibilities of increased pill burden if the patient is on Anti-Retroviral Treatment (ART), increased incidences of side effects and other co-morbid conditions which result from the HIV such as depression and dementia, may increase the likelihood of poor compliance. 2.3 ENVIRONMENTAL RELATED FACTORS FOR DEFAULT AMONG TB PATIENTS According to some studies done on default among TB patients, the following environmental related factors were discussed: health system factors, compliance and non-compliance, stigma and discrimination. Health System factors will include correlation between patient and University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 12 programme needs, structural factors and staff knowledge and attitudes. In depth interviews during a study in India, to “assess the needs and perspectives of patients and providers in two chest clinics in Delhi”, showed that reasons for default are linked to poor correlation between patient and programme needs or priorities, and to particular characteristics of disease and its treatment. Some of the patient needs that are still to be met by health systems are convenient clinic timings, arrangements for provision of treatment in family emergencies, and provision for complicated cases such as alcoholics (Jaiswal et al 2003:625). In a study in Nepal inconvenient opening times for TB Clinics situated far from patients’ homes accounted for defaulting in 28% of non-compliant TB patients (Bam et al 2005:54). Both studies recommended flexible clinic opening times to accommodate patients staying at a distance, so as to improve compliance to treatment. A systematic review of qualitative research on patient adherence to TB treatment identified the barriers to completing TB treatment as structural (including poverty and gender discrimination), social, health service-related and personal (Munro et al 2007:238). A number of studies have shown the importance of the relationship between health care workers and the patients as contributing to treatment compliance or non-compliance. Bam et al (2005:55), in a study conducted in Nepal, found that the quality of the health care provider and patient interaction and relationship contributed to differences in treatment adherence. Similarly, a study in South Africa (Peltzer et all 2002:67) established that the quality of healthcare provider and patient communication, coupled with correct causative belief, were associated with TB treatment compliance. Jaiswal et al (2003:625) demonstrated in a study in India that problems facing patients were poor interpersonal communication with health staff, lack of attention and support at the clinic, difficulties for patients to re-enter the system if they missed treatment, and long distances to the health facilities (inaccessibility). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 13 The disease related problems were inability of the staff to manage side effects of medicine and patients perception of equating feeling well or better with being cured. A Madagascar study also identified quality of relationships of patients with medical staff and staff knowledge and attitudes regarding TB as contributing to compliance or non-compliance to TB treatment (Comolet et al 1998). In this study, quality of relationships with and attitudes of healthcare workers had a significant bearing on TB treatment compliance, as they determined whether patient would return for treatments in the facility. In this study, compliance occurs when the patient take TB treatments without interruption for more than two months and completes the course. Non-compliance occurs when a patient interrupts treatment for more than two months. Compliance to TB treatment is one of the most important factors that determine the outcome of treatment, and the extent to which a patient’s behaviour while on TB treatment coincides with medical advice (Pandit et al 2006:241). Non-compliance to TB treatment has been associated with a number of factors as described. A study conducted in Colombo showed treatment compliance was negatively affected by disease-related problems such as inability of the staff to manage medicine side effects and patients’ perception of equating feeling well or better with being cured (Janakan et al 2008:214-23). The study suggests that delays in addressing medicines’ adverse effects or ignoring patients’ complaints about adverse effects of medicines they are taking may thus promote non-compliance to treatment (Janakam et al 2008:222). A Thailand study also demonstrated an association of poor compliance to TB treatment and adverse medicinal effects (Okanurak et al 2008:1160). According to the WHO (2008a:3), most TB patients complete treatment without experiencing significant side effects from taking the TB medicines. The few patients who University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 14 report or develop side effects commonly present with skin rashes, visual and auditory disturbances, jaundice, burning sensations in the limbs and painful limbs. These side effects or fear of side effects may cause patients to comply poorly with their treatment. Therefore, attention should be paid to diagnosis and prompt treatment of side effects, as well as educating patients about the possible side effects. According to Kudakwashe:2010), some of the barriers postulated to be contributing to poor TB treatment compliance are:  Communication difficulties  Low literacy levels  Inadequate knowledge and low awareness of TB disease  Patient attitudes and beliefs in treatment efficacy  Depression and other psychiatric illnesses  Alcohol and substance abuse  Unstable living conditions  Negative health provider attitudes  Stigma and discrimination  Overcrowding and access to medicines. The presence or perceived presence of stigma and discrimination in a community members, who may provide the much needed psychosocial support to the patient (Eastwood et al University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 15 2004:70). The same authors in a Gambian qualitative study to explore gender differences in care seeking behaviour demonstrated that patients often consulted traditional healers initially due to perceive stigma from health workers and the community. The study interviewed health workers and TB patients (male and female). Stigma therefore may result in delays in seeking treatment or taking treatment consistently and correctly. Furber et al (2004:1281) also concluded that stigma and discrimination of TB and HIV patients delaying seeking testing and treatment and thus poorer health outcomes. In most African societies, TB and HIV are associated with immoral behaviour and patients suffering from these conditions would be hesitant to disclose their status to their family members, a situation which may result in these patients not complying with their treatments as they do not want to be seen taking the medicines (Kaona et al 2004:68). 2.4 STRATEGIES USED TO CONTROL TUBERCULOSIS The World Health Assembly (WHA) passed a resolution that recognized TB as a major global public health problem followed by the launch of DOTS as the internationally recommended TB control strategy in 1994 (WHO 2006b:6). Most National TB Control Programmes (NTCP) experienced major progress in TB control when they implemented the DOTS strategy (WHO 2006b:6). The DOTS strategy was later expanded to form the Stop TB strategy, which seeks to build on the successes of the DOTS strategy (WHO 2006b:5), and which will be discussed as a TB control strategy. Most of the successes in TB control globally have been attributed to the DOTS strategy, especially in high TB burden countries (WHO 2006b:6). In order to address the remaining challenges, especially in areas where the TB epidemic has been worsening (such as Sub- Saharan Africa and Eastern Europe), and to achieve the Millennium Development Goal University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 16 (MDG) and related Stop TB partnership targets by 2015, a new strategy was developed, namely the Stop TB Strategy (WHO 2006b:6; Dye et al 2005:2767). According to WHO (2006c:9) with DOTS as the central component, the Stop TB strategy set out-steps which national TB control programmes, their partners and state-holders needed to take to improve TB control. To achieve TB control requires a comprehensive and persistent response that complements other measures aimed at addressing social and environmental factors that increase the risk of individuals to develop TB (WHO 2006b:11). Thus, further DOTS strengthening is required in the following areas, namely: political commitment with increased and sustained financing; case detection through quality assured bacteriology; standardized treatment with supervision and support; effective drug supply and management systems; monitoring and evaluation systems; and impact assessment (WHO 2006b:9). The HIV epidemic has worsened the global burden of TB and increased the need to focus attention on strengthening the global TB and HIV programmes in order to tackle the two public health problems effectively (WHO 2004:3). TB has become the leading cause of death among people living with HIV, while infection with HIV is a large risk factor risk factor for latent and recent TB infection to convert to active TB disease (WHO 2006c:24). The international standards for TB/HIV as setout by WHO are aimed firstly at decreasing the burden of TB among people living with HIV by strengthening intensive TB case finding, provision of Isoniazid preventive therapy (IPT) for TB/HIV co-infected patients and TB infection control in healthcare and congregate settings. Secondly, the standards aim at decreasing the burden of HIV prevention and Cotrimoxazole prophylaxis, and HIV care and support, including provision of ART for eligible patients (WHO 2004:2). Therefore, collaborative activities between the TB and HIV programmes using the above strategies University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 17 should be implemented as they can help control TB among HIV patients (Maher et al 2002:3; WHO, 2006b:1). The WHO (2006d:12) describes MDR TB as a threat to global TB control, worsened by inadequate treatment for those suffering from it; increase in MDR TB patients due to misuse of second line anti TB medicines; and absence of new effective anti-medicines. NTCPs should ensure early detection and correct treatment of all forms of DR-TB and patient adherence to this treatment. TB control programmes also need to pay attention to special population groups such as prisoners, refugees and other high-risk groups associated with high TB transmissions due to overcrowding and poverty (WHO 2006b:13). Improving access to quality health care services will benefit TB control, therefore TB control programmes should actively improve system wide policy, human resources, financing, management, service delivery and information systems (WHO 2006b:13). Many patients with early symptoms of TB do actually consult private health care providers first and many such providers diagnose and treat TB (WHO 2006b: 14). The diagnosis of TB needs to be made without delay and once done the right treatment with adequate dosing need to be instituted with proper follow up of such patients. Thus, engagement of all health care providers (both private and public) is of paramount importance. Evidence suggests that failure to engage all care providers used by TB suspects and patients hampers TB case detection, delays diagnosis, leads to incorrect diagnosis as well as inappropriate and in complete treatment. Increases drug resistance and places unnecessary financial burden on the patients and health systems (Uplekar et al 2001:912). In order to standardize the diagnosis and treatment of TB both in the public and private University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 18 sector, the Tuberculosis Coalition for Technical Assistance (TBCTA) developed a tool known as the International Standards for Tuberculosis Care (ISTC) to guide these sectors (WHO 2006c:6). The tool was designed with the idea of having the health care provider at the centre of TB control activities and the patient at the centre of care (WHO 2006b:15). The document has seventeen standards of care which are grouped into six standards of diagnosis, nine standards for treatment and two for addressing public health responsibilities (Migliori et al 2006:687). It places special emphasis on prompt diagnosis of TB with bacteriological examination where possible, correct treatment at the right doses and monitoring treatment with bacteriological examination, as well as evaluation of close contacts (TBCTA 2006:5). To achieve the maximum use of this tool there has to be collaboration with and from the NTCPs, with a deliberate attempt to promote public private mix using the DOTS approaches. To achieve greater commitment to fight TB, Advocacy Communication and Social Mobilisation (ACSM) embraces the following: advocacy to influence policy changes and ensure sustained financial and political commitment; facilitation of communication between health care providers, TB patients and their communities in order to improve knowledge of TB and subsequently compliance to treatment; and social mobilization to engage the communities, partners and stakeholders in the fight against TB (WHO 2006b:15). Conducting locally relevant operational research can identify challenges and practical solutions that can be tested in the field before scaling up the activities (WHO 2006b:16). NTCPs can thus develop new and effective strategies for TB control. The WHO (2006b:16) advocates TB programmes to facilitate and actively support research to develop new diagnostics, drugs and vaccines. Studies in Malaysia and Zambia showed that non-compliance was associated with being free of symptoms (O’Boyle et al 2002:307; Kaona et al 2004:68). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 19 Patient’s defaulting behaviour occurs when the TB symptoms disappear and they feel well, usually after a few months of TB treatment (Pushpananthan et al 2000:216; Peltzer et al 2002:55). Therefore, according to Bam et al (2005:56), the authors conclude that compliance to TB treatment could be improved by providing more information about TB to the patients. Compliance to TB treatment can, therefore, be improved by promoting TB treatment literacy among those with the disease, their families and communities. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 20 CHAPTER 3 METHODS 3.0 STUDY SETTING The study site for this research was Freetown the capital city of Sierra Leone. Freetown has a population more than one (1) million inhabitants, the chief port and the largest city of Sierra Leone. It is located on the rocky Sierra Leone peninsula on the Atlantic coast. Freetown was founded by Britain as a settlement for freed slaves in 1787. It was made capital of the independent Sierra Leone in 1961. Freetown is of tropical climate with a rainy season - May through October, the rest of the year representing the dry season. The beginning and end of the rainy season is marked by strong thunder storms. This is a tropical savannah climate. 3.1 STUDY DESIGN The study design for this study was an unmatched case control study. 3.2 STUDY POPULATION The study population was tuberculosis patients registered at the Connaught Chest Clinic in Western Sierra Leone. The population consisted of tuberculosis patients on treatment who defaulted (non-compliant) and those who have completed treatment (compliant). The study population was comprised of the cohort of patients registered during the period September 2012 to August 2013 at the Connaught Chest Clinic. Cases were patients who interrupted treatment for two consecutive months or more (as defined by WHO), and controls were patients who completed the treatment course of six months. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 21 Inclusion criteria for this study was all smear microscopy cases registered at the Connaught Chest Clinic during the study period and must have completed treatment or default during the study period. 3.3 SAMPLE SIZE The TB facility register of the Connaught Chest Clinic was used to establish the sampling frame. The register has records of the registered patient’s name, registration number, date of starting, treatment outcome, HIV status, demographic details, address, and classification of diagnosis. The sample size for this study was Fleiss unmatched case control sample size method. Two sided confidence level (1 - alpha) 95 Power (% chance of detecting) 80 Ratio of control to cases 1 Hypothetical proportion of controls with exposure 12 Hypothetical proportion of cases with exposure 28.34 Least extreme Odds Ratio to be detected 2.9 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 22 Fleiss Fleiss CC Sample size-cases 94 106 Sample size-controls 94 106 Total sample size 198 212 The Fleiss continuity correction (cc) was used as the sample size for this research. However, 120 cases and 120 controls was used for this study. 3.4 SAMPLING PROCEDURE From the Connaught chest clinic facility register, all patients who defaulted and completed treatment within the study period was enrolled. A sample frame was formed with using all patients who defaulted and completed treatment within the study period .The participants for this study was selected by simple random sampling method from the sampling frame. 3.5 STUDY PROCEDURE A structured questionnaire was used to capture data from individuals in the study population, and was comprised of closed ended questions that makes it easier to analyze the data. The questionnaire was divided into seven section (A to G), covering demographic information, patient related factors, socio-economic variables, health care system-related factors, default factors, stigma and discrimination, disease and medicine-related factors. Responses to each question was coded. A coded box, where the coding number for each of the responses to be entered was inserted for each question in the questionnaire. A list of variables was arrived at, after a review of relevant existing literature. The questionnaire was administered by research assistants who are familiar with the geography of Freetown and fluent in the local language, University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 23 but who had not worked at the Connaught Chest Clinic. A pilot test of the questionnaire was done at the TB clinic and appropriate adjustment was made. After the sampling process was completed, a list of possible study participants was created using their names and contact details. The Research Assistants (RA) visited the participants at their respective homes. The RA introduced themselves to the prospective participants and read through the individual participant consent from that detailed the title and purpose of the research as well as the rights of the participants and details of the person to be contacted for future questions. When a participant agrees to be interviewed, he/she was asked to provide written consent by signing or thumb printing. After obtaining the written consent, the RA entered the questionnaire serial number and date of interview and proceed with the interview using a language understandable by the participant. The RA entered responses given by the participant by circling the appropriate response number and enter the same number into the coding box. This was done to ensure data quality as the response number circle suppose to be the same as the one entered in the coding box. The process of data collection was continued until every effort to contact every study participant in the sample was exhausted. All completed questionnaire was kept in a safe cupboard with keys. Secondary data for each participant was captured from the facility register. In order to achieve the required sample size, replacement was made using the sample frame for those that were not traced or dead. 3.6 ETHICAL APPROVAL Ethical approval for this study was obtained from the Ethical Review Committee in Sierra Leone, in the Ministry of Health and Sanitation. TB is strongly associated with HIV/AIDS University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 24 leading to stigmatization and is difficult to discuss in public. Informed consent was obtained from patients and confidentiality and anonymity was assured. Names and addresses of patients was collected only for the purposes of follow-up. 3.7 DATA ANALYSIS All questionnaires were reviewed by the researcher for completeness and appropriate data cleaning was done. Data was then entered on a spreadsheet using MS Excel. Descriptive summary statistics and graphical summaries in charts are presented. Data was analysed using logistic regression, bivariate and multivariate analysis. 3.8 LIMITATION OF THE STUDY The limitations of this study were that some member of the study population have transferred from their previous addresses and some have died. In order to achieve the required sample size for this study replacement was made using the sample frame for those who were not traced or dead by simple random sampling method. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 25 CHAPTER 4 RESULTS 4.1 DEMOGRAPHIC FACTORS 4.1.1 RESPONDENTS’ GENDER (N : 240) Of all the 240 respondents, 63.3% (152) were males and 36.7% (88) females. Though no significant gender association occurred between the compliant and non-compliant groups(Table 4.1). There was a higher proportion of males in the both categories.There were 62.5% (75) males and 37.5% (45) females in the complaint group, whilst in the non- compliant group males accounted for 64.2% (77) and females 35.8% (43) of the total. The gender distribution among the cases and controls were as shown in figure 4.1 (below). Figure 4.1 Sex distribution of cases and controls of TB treatment compliance 25 CHAPTER 4 RESULTS 4.1 DEMOGRAPHIC FACTORS 4.1.1 RESPONDENTS’ GENDER (N : 240) Of all the 240 respondents, 63.3% (152) were males and 36.7% (88) females. Though no significant gender association occurred between the compliant and non-compliant groups(Table 4.1). There was a higher proportion of males in the both categories.There were 62.5% (75) males and 37.5% (45) females in the complaint group, whilst in the non- compliant group males accounted for 64.2% (77) and females 35.8% (43) of the total. The gender distribution among the cases and controls were as shown in figure 4.1 (below). Figure 4.1 Sex distribution of cases and controls of TB treatment compliance 25 CHAPTER 4 RESULTS 4.1 DEMOGRAPHIC FACTORS 4.1.1 RESPONDENTS’ GENDER (N : 240) Of all the 240 respondents, 63.3% (152) were males and 36.7% (88) females. Though no significant gender association occurred between the compliant and non-compliant groups(Table 4.1). There was a higher proportion of males in the both categories.There were 62.5% (75) males and 37.5% (45) females in the complaint group, whilst in the non- compliant group males accounted for 64.2% (77) and females 35.8% (43) of the total. The gender distribution among the cases and controls were as shown in figure 4.1 (below). Figure 4.1 Sex distribution of cases and controls of TB treatment compliance University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 26 4.1.2 AGE OF RESPONDENTS (N=240) The questionnaire grouped the respondents into age groups. The age distributions are shown in Figure 4.2. Figure 4.2Age distribution of case and control of TB treatment completion Figure 4.2 shows that 57.5% (138) of respondents were between 25 and 44 years of age. The younger age groups between 25 and 34 years, accounted for both the highest proportions of compliant (39%) and non-compliant (39%) respondents. Age distribution did not account for any significant difference between the two groups (Table 4.1). 4.1.3 RESPONDENTS’ RELIGION (N = 240) A total of 35.8% (86) respondents reported they were Christians and 64.2% (154) reported they were Islam. Christians were equally distributed in both the compliant and non-compliant groups. The respondents’ distribution with respect to their religion was as shown in Figure 4.3. All respondent’s belonged to either of the two major religious groups in the country. 26 4.1.2 AGE OF RESPONDENTS (N=240) The questionnaire grouped the respondents into age groups. The age distributions are shown in Figure 4.2. Figure 4.2Age distribution of case and control of TB treatment completion Figure 4.2 shows that 57.5% (138) of respondents were between 25 and 44 years of age. The younger age groups between 25 and 34 years, accounted for both the highest proportions of compliant (39%) and non-compliant (39%) respondents. Age distribution did not account for any significant difference between the two groups (Table 4.1). 4.1.3 RESPONDENTS’ RELIGION (N = 240) A total of 35.8% (86) respondents reported they were Christians and 64.2% (154) reported they were Islam. Christians were equally distributed in both the compliant and non-compliant groups. The respondents’ distribution with respect to their religion was as shown in Figure 4.3. All respondent’s belonged to either of the two major religious groups in the country. 26 4.1.2 AGE OF RESPONDENTS (N=240) The questionnaire grouped the respondents into age groups. The age distributions are shown in Figure 4.2. Figure 4.2Age distribution of case and control of TB treatment completion Figure 4.2 shows that 57.5% (138) of respondents were between 25 and 44 years of age. The younger age groups between 25 and 34 years, accounted for both the highest proportions of compliant (39%) and non-compliant (39%) respondents. Age distribution did not account for any significant difference between the two groups (Table 4.1). 4.1.3 RESPONDENTS’ RELIGION (N = 240) A total of 35.8% (86) respondents reported they were Christians and 64.2% (154) reported they were Islam. Christians were equally distributed in both the compliant and non-compliant groups. The respondents’ distribution with respect to their religion was as shown in Figure 4.3. All respondent’s belonged to either of the two major religious groups in the country. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 27 The religion practical did not account for any significant difference between two compliant and non- compliant groups (Table 4.1). Figure 4.3Religion distribution of case and control of TB treatment completion 4.1.4 MARITAL STATUS (N-240) With reference to marital status, 32.9% (79) of respondents reported that they were married, 103 (44.9%) of respondents were single. The largest proportions of respondents in both compliant and non-compliant group were found among the single and married respondents (Figure 4.4). Furthermore, there were higher levels of compliance among both married 54.4%(43) and single 53.4% (55) respondents, compared to the non-compliant group. Only 24.2% (58) of the 240 respondents were widowed, separated and cohabiting. However, marital status did not account for any significance difference between the complaint and non- compliant groups (Table 4.1). 27 The religion practical did not account for any significant difference between two compliant and non- compliant groups (Table 4.1). Figure 4.3Religion distribution of case and control of TB treatment completion 4.1.4 MARITAL STATUS (N-240) With reference to marital status, 32.9% (79) of respondents reported that they were married, 103 (44.9%) of respondents were single. The largest proportions of respondents in both compliant and non-compliant group were found among the single and married respondents (Figure 4.4). Furthermore, there were higher levels of compliance among both married 54.4%(43) and single 53.4% (55) respondents, compared to the non-compliant group. Only 24.2% (58) of the 240 respondents were widowed, separated and cohabiting. However, marital status did not account for any significance difference between the complaint and non- compliant groups (Table 4.1). 27 The religion practical did not account for any significant difference between two compliant and non- compliant groups (Table 4.1). Figure 4.3Religion distribution of case and control of TB treatment completion 4.1.4 MARITAL STATUS (N-240) With reference to marital status, 32.9% (79) of respondents reported that they were married, 103 (44.9%) of respondents were single. The largest proportions of respondents in both compliant and non-compliant group were found among the single and married respondents (Figure 4.4). Furthermore, there were higher levels of compliance among both married 54.4%(43) and single 53.4% (55) respondents, compared to the non-compliant group. Only 24.2% (58) of the 240 respondents were widowed, separated and cohabiting. However, marital status did not account for any significance difference between the complaint and non- compliant groups (Table 4.1). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 28 Figure 4.4 Marital status of case and control to TB treatment completion 4.1.5 EDUCATIONAL LEVEL ATTAINED (N = 240) Among the respondents 44.2% (106) had secondary education, 34.6% (83) no formal education, 15.4% (37) primary education and 5.8% (14) tertiary education. About 54% of the respondent with secondary education did not complete TB treatment; similarly 56.7% (21 out of 37) respondents with primary education,46.9% (39 of 83) with no formal education and 21.4% (3 of 14) with tertiary education did not complete their TB treatment (Figure 4.5). The level of education attained was not a significant factor in determining compliance and non-compliance to TB treatment (Table 4.1). Figure 4.5 Educational levels attained case and control of TB treatment completion 28 Figure 4.4 Marital status of case and control to TB treatment completion 4.1.5 EDUCATIONAL LEVEL ATTAINED (N = 240) Among the respondents 44.2% (106) had secondary education, 34.6% (83) no formal education, 15.4% (37) primary education and 5.8% (14) tertiary education. About 54% of the respondent with secondary education did not complete TB treatment; similarly 56.7% (21 out of 37) respondents with primary education,46.9% (39 of 83) with no formal education and 21.4% (3 of 14) with tertiary education did not complete their TB treatment (Figure 4.5). The level of education attained was not a significant factor in determining compliance and non-compliance to TB treatment (Table 4.1). Figure 4.5 Educational levels attained case and control of TB treatment completion 28 Figure 4.4 Marital status of case and control to TB treatment completion 4.1.5 EDUCATIONAL LEVEL ATTAINED (N = 240) Among the respondents 44.2% (106) had secondary education, 34.6% (83) no formal education, 15.4% (37) primary education and 5.8% (14) tertiary education. About 54% of the respondent with secondary education did not complete TB treatment; similarly 56.7% (21 out of 37) respondents with primary education,46.9% (39 of 83) with no formal education and 21.4% (3 of 14) with tertiary education did not complete their TB treatment (Figure 4.5). The level of education attained was not a significant factor in determining compliance and non-compliance to TB treatment (Table 4.1). Figure 4.5 Educational levels attained case and control of TB treatment completion University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 29 Table 4. 1 Demographic factors for case and control Factors Case N(%) Control N(%) OR(95 % CI) P Value Gender 0.788 Male 77 (64.2) 75 (62.5) 1 Female 43 (35.8) 45 (37.5) 0.9(0.55-1.57) 0.789 Age groups (years) 0.064 15 - 24 20 (16.7) 32 (26.7) 1 25 - 34 39 (32.5) 39 (32.5) 1.6(0.78-3.26) 0.197 35 - 44 38 (31.7) 22 (18.3) 2.7(1.28-5.95) 0.009 45 - 54 19 (15.8) 19 (15.8) 1.6(068-3.73) 0.276 55 - 64 2 (1.7) 7 (5.8) 0.4(0.08-2.24) 0.356 15 - 24 2 (1.7) 1 (0.8) 3.2(0.27-37.62) 0.355 Religion 0.105 Christian 49 (40.8) 37 (30.8) 1 Islam 71 (59.2) 83 (69.2) 0.64(0.37-1.09) 0.107 Marital Status 0.052 Married 36 (30) 43 (35.8) 1 Widowed 8 (6.7) 11 (13.3) 0.8(0.31-2.39) 0.785 Separated 17 (14.2) 6 (5) 3.4(1.21-9.48) 0.020 Single 48 (40) 55 (45.8) 1.0(0.57-1.87) 0.890 Cohabiting 11 (9.2) 5 (4.2) 2.6(0.83-8.26) 0.098 4.2 PATIENT-RELATED FACTORS 4.2.1 CIGARETTES SMOKING (N=240) A total of 17.9% (43) of the respondents reported having smoked cigarettes during the previous six months while 82.1% (197) had not done so (Fig 4.6). There were a higher proportion of respondents who had smoked in the previous six months in the non-compliant group (26.7%) compared with the compliant group (9.2%). Smoking cigarettes in the previous six months was found to be significantly associated with compliant or non- compliant to TB treatment groups (OR=0.3(0.14-0.63))(Table 4.2). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 30 4.2.2 DRINKING ALCOHOL (N =240) A total of 19.6% (47) drank alcohol in the previous six months while 80.4% (193) had not done so. Among those who had drank alcohol, 26.7% (32) did not complete TB treatment (non-compliant), compared to 73.3% (88) patients in the same non-compliant group who had not drunk any alcohol in the previous six months. There was significant association between whether or not the respondent had drunk alcohol in the previous six months and their TB treatment compliance status(OR=0.3(0.19-0.77)) (Table 4.2). 4.2.3 AVAILABILITY OF TREATMENT SUPPORTER (N = 240) Among respondents, 74.2% (178) of the patients had treatment supporters while they were taking treatment and only 25.8% (62) of the respondents did not have one. However 61.8% (110) of the respondent who had treatment supporter were compliants while 68 (38.2%) were non-compliant. The analysis showed that having a treatment supporter was significantly associated with compliant to TB treatment, indicating that respondents with having treatment supporters are most likely to complete their TB treatment (OR=8.4(4.0-17.6)) (Table 4.2). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 31 Table 4. 2 Patient related factor for case and control Factors Case N(%) Control N(%) OR(95%CI) Pvalue Smoking 0.0007 Yes 32 (26.7) 11 (9.2) 1 No 88 (73.7) 109 (90.8) 0.3(0.14-0.63) 0.001 Alcohol 0.005 Yes 32(26.7) 15(12.5) 1 No 88(73.3) 105(87.5) 0.3(0.19-0.77) 0.007 Treatment Supporter 0.000 Yes 68(56.7) 110(91.7) 1 No 52(43.3) 10(8.3) 8.4(4.0-17.6) 4.3 SOCIO-ECONOMIC FACTORS 4.3.1 EMPLOYMENT STATUS (N=240) Of the 240 respondents, 50.8%(122) unemployed, 30.4% (73) self-employed and only 18.8% (45) were employed.Respondents in all three employment categories were fairly distributed among the compliant or non-compliant to TB treatment completion groups (Fig. 4.6).Employment status was not a significant factor in determining compliance or non- compliance to TB treatment completion (Table 4.3). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 32 Figure 4. 6 Employment status of case and control of TB treatment completion 4.3.2 INCOME LEVEL OF RESPONDENTS Whether or not respondent employed, self-employed or unemployed they were asked a follow up question to determined their level of income per month in the Leones (Le). The income level divided into three groups: low (Le1,000,000.00). Among the respondents 86.7% (208) were low income earners, 12.9% (31) medium and 0.4% (1) high income earner (Fig 4.7). However, the level of income was not associated with any significant difference between the compliant and non-compliant groups (Table 4.3), because these income categories were fairly equally distributed in both groups (Fig 4.7). 0 5 10 15 20 25 30 35 Employed Self-employed Unemployed P er ce nt ag e Respondents' employment status CASE CONTROL University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 33 Figure 4.7 Distribution of income level s among case and control of TB treatment completion 4.3.3 FOOD AVAILABILITY AMONG RESPONDENTS (N=240) On the question of food availability among respondents, 75.4% (181) reported that food was always available, 17.1% (41) reported that food was available most of the time, and 7.5% (18) said food was not always available. A total of 88.9% (16) respondents who reported that they did not always have food to take with medicines, subsequently did not complete their TB treatment, while 11.1% (2) who also did not always have food nevertheless completed their treatment. The availability of food was found to be a significant factor associated with difference between compliant and non-compliant groups to TB treatment completion (OR=12(2.83-56.85)) (Table 4.3). 0 5 10 15 20 25 30 35 40 45 50 Low Medium High P er ce nt ag e Level of income CASE CONTROL University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 34 Table 4. 3 Socio- economic factors for case and control respondents Factors Case N(%) Control N(%) OR(95% CI) PValue Employment status 0.061 Employed 18 (15) 27 (22.5) 1 Self-employed 32 (26.7) 41 (34.2) 1.1(0.55-2.48) 0.682 Unemployed 70 (58.3) 52 (43.3) 2.0(1.0-4.0) 0.048 Income level (Le) 0.8669 < 400,000 104 (86.7) 104 (86.7) 1 400,000 – 1,000,000 16 (13.3) 16 (13.3) 1.0(0.50-2.26) 0.867 >1,000,000 0 1 (0.8) Food availability Always available 70 (58.3) 111 (92.5) 1 Available most times 34 (28.3) 7 (5.8) 7.7(3.23-18.32) 0.000 Not always available 16 (23.3) 2 (1.7) 12.6(2.83-56.85) 0.001 4.4 HEALTH CARE SYSTEM - RELATED FACTORS 4.4.1 CONVENIENT TB CLINIC OPENING TIMES (N=240) Respondents response to which TB clinic opening times were most convenient for them show that 80.4% (193) said that before 0.8hrs - 17hrs would be the most convenient opening times for the TB Clinic, followed by 17.5%(42) who wanted the clinic to be open at 08hrs until 17hrs, and 2.1%(5) to open at 0.8 hrs and close after 17 hrs (Fig 4.8). There was a significant association difference between preference for TB clinic opening time and the compliant and non-compliant to TB treatment completion (p=0.003) (Table 4.4). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 35 Figure 4. 8 Respondents' preference for TB clinic times 4.4.2 WAITING TIME AT CLINIC (N=240) Respondents were asked how long they usually waited at the TB Clinic before they were attended. 41.3% (99) of the respondents said they usually waited for less than an hour at the TB clinic before they were attended to, while 47.1%(113) had to wait between one and two hours and 11.6% (28) usually waited for more than two hours. Waiting time at clinic was significantly associated with respondents’ compliance and non-complianceto TB treatment completion(P=0.000) (Table 4.4). This means that longer waiting times were associated with non-compliance to completion of TB treatment, as indicated by respondents. 4.4.3 DISTANCE TRAVELLED (N = 240) The respondents were asked to indicate how much distance they travelled to go and pick up their medicines. 42.1% (101) travelled less than five kilometers while 35% (84) between five and ten kilometers, 14.6% (35) between eleven and fifteen kilometers, 5% (12) between sixteen and twenty kilometers and 3.3% (8) still had to travel more than twenty 0 20 40 60 80 100 120 <8am-5pm 8am-5pm 8am->5pm <8am->5pm N um be r of R es po nd en t Clinic times CASE CONTROL University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 36 kilometers.Distance was not a significant factor between the compliant and non-compliant groups to TB treatment completion (Table 4.4). This may be explained by the much lower proportion of respondents that travelled more than 15 km to TB clinics, which was equally distributed between the two groups. 4.4.4 COST OF TRANSPORT (N = 240) The cost of transportation to TB clinic was tested as a factor influencing respondents completion of treatment. In total, 40.8% (98) respondents did not have to pay anything, 52.5% (126) had to pay less than Le 10,000 and 6.7% (16) paid between Le10,000 and Le15,000 each time they went to the clinic, 38.3% (46) of the respondents in the non- compliant group did not have to pay anything to get to the clinic, while 2.5% (15) had to pay between Le10,000.00 and Le15,000.00. There was significant difference between the compliant and non-compliant groups with respect to the cost of getting to the clinic to collect medicines those paying between Le 10,000-Le 15,000(Table 4.4) 4.4.5 DOT STATUS (N = 240) The respondents were asked who was supervising them while they had to take their TB medicines. The options they had to choose were either health worker, family member, community member or none. The DOT status was observed for the first two months at the clinic. 2.1% (5) had no one supervising them, 1.3%(3) were surpervised by family members and 96.7% (232) by health workers. DOT status did not contribute to any significant difference between the two groups. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 37 4.4.6 HEALTH WORKERS ATTITUDES TOWARDS PATIENTS (N= 240) The respondents rated the attitude of health workers who attended them at the TB clinic as follows: 69.2% (166) very friendly, 25.4% (61)Friendly, 5% (12) unfriendly and 0.4% (1) very unfriendly. The only patient who rated the attitude of the health worker who attended to them at the clinic as unfriendly completed his/her TB treatment.There was no significant association between health workers attitude and compliant or non-compliant to TB treatment completion (Table 4.4). 4.4.7 MEDICINE AVAILABILITY (N = 240) Respondents were asked about the availability of medicines when they attended TB clinic. 70.4% (169) said medicine was always availability,33.1% (56) of those whose said medicine was always available did not complete TB treatment. 29.6% (71) of the respondents said that medicine was sometimes available, 90.1% (64) of them did not complete their treatment. The availability of medicine was found to have a significant association between compliance or non-compliance to TB treatment completion (OR = 18.4(7.93-42.88)) (Table 4.4). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 38 Table 4. 4 Healthcare system related factors for case and control respondents Table 4.4 Health care related system for case and control Factors Case N(%) Control N(%) OR(95% CI) PValue TB Clinic Opening Times 0.003 Before 8 am - 5pm 86 (71.7) 107 (89.2) 1 8 am - 5 pm 29 (24.2) 13 (10.8) 2.7(1.36-5.66) 0.005 8 am - later than 5pm 5 (4.2) 0 Clinic Waiting Times 0.0000 < 1 HR 46 (38.8) 53 (44.2) 1 2 HRS 70 (58.3) 43 (35.8) 1.87(1.08-3.24) 0.024 > 2 HRS 4 (3,3,) 24 (20) 0.2(0.06-0.59) 0.004 Distance to the clinic(km) < 5 70 (58.3) 56 (46.7) 1 5 - 10 34 (28.3) 40 (33.3) 1.7(0.76-2.44) 0.289 10 - 15 25 (20.8) 10 (8.3) 3.1(1.35-7.14) 0.007 16 - 20 6 (5) 6 (5) 1.2(0.37-4.12) 0.720 > 20 0 8 (6.7) Travel cost (x Le 1000) 0 42 (35) 56 (46.7) 1 <10 63 (52.5) 63 (52.5) 1.33(0.78-2.26) 0.288 10 –15 15 (12.5) 1 (0.8) 20(2.54-157.45) 0.004 Attitude of health workers Friendly 39 (32.5) 22 (18.3) 1 Very Friendly 70 (58.3) 96 (80) 2.4(1.32-4.55) 0.004 Unfriendly 11 (9,2) 1 (0.8) 15.1(1.90-119.57) 0.01 Very unfriendly 0 1 (0.8) Avaliability of drugs Always 56 (46.7) 113 (94.2) 1 Sometimes 84 (53.3) 7 (5.8) 18.4(7.93-42.88) 0.000 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 39 4.5 KNOWLEDGE OF TUBERCULOSIS AND TREATMENT 4.5.1 KNOWLEDGE OF SYMPTONS (N=240) Knowledge of the symptoms asked about to TB respondents included coughing, night sweats, loss of weight and chest pain, to which they responded with a “yes” or “no” respectively. In total, over 50%of compliant and non-compliant groups, respectively,had knowledge of the symptoms for TB. Of the respondents 95% (228) said coughing as a symptom,81.7% (196) said chest pain, 63.8% (153) said night sweat and 76.7% (184) said loss of weight are symptoms of TB. These proportions were distributed equally between the compliant and non- compliant groups and so there was no significant association between knowledge of symptoms and compliance or non-compliance to completion of TB treatment (Table 4.5). 4.5.2 PERIOD OF TREATMENT FOR TB (N = 240) The questions were whether TB should be treated for six months, until one feels better or six months and health worker stops them. With regards to treatment should be for six months, 87.1% (209) said “yes” and 12.9% (31) said “no”. All the eight respondents who said “ no” did not complete their treatment in addition to the 112 of the 209 who said “yes”. Perception on the period for treatment was found to be significantly associated with completion of TB treatment (P= 0.003),(Table 4.5). Respondents who think TB treatment should be taken until one feels better accounted for higher proportion of non-compliance to TB treatment completion. Also, there was no statistical association between the compliant and non- compliant groups(P=0.08 (Table 4.5). Respondents who thinks one should take the treatment for six months and health worker tells them to stop was found to be statistically associated with the completion of TB treatment (P= 0.001) (Table 4.5). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 40 Table 4. 5 Knowledge and treatment of TB for case and control respondents 4.6 DEFAULT FACTORS 4.6.1 WHO RESPONDENTS LIVES WITH (N = 240) Respondents were interviewed in respect to who they live with 79.6% (191) lived with their family, 12.5% (30) friends and 7.9% (19) alone. Among the respondent who lived with their family, 45% (86) were in the non-compliant group. Also among the 19 respondents who lived alone 78.9% (15) were in the non-compliant group (Fig. 4.9). There was a strong significant association between who the respondent lives with and the compliance status to TB treatment completion (P= 0.004) (Table 4.6). Factors Case N(%) Control N(%) OR(95% CI) P Value Treatment for 6 months or more 0.003 YES 112 (93.3) 97 (80..8) 1 NO 8 (6.7) 23 (19.2) 0.3(0.12-0.70) 0.006 Treatment until feels better 0.08 YES 7 (5.8) 2 (1.7) 1 NO 113 (94.2) 118 (98.3) 0.3(0.05-1.34) 0.111 Treatment until health worker tells you to stop 0.001 YES 118 (98.3) 106 (88.3) 1 NO 2 (1.7) 14 (11.7) 0.1(0.03-0.57) 0.007 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 41 Figure 4. 9 A distribution graph of the person case and control live with 4.6.2 HOW MANY OTHER PEOPLE DO YOU LIVE WITH (N = 240) Of the 240 respondents, 7.9% (19) live alone, >80% (15) of which are in the non-compliant group; 25% (60) lived with one to three others, 41.7% (100) lived with four or six others and 25.4%(61) lived with others greater than seven. The number of people respondents live with was not a significant factor associated with compliance or non-compliance to TB treatment completion (Table 4.6). 0 20 40 60 80 100 120 Family Friends Alone N um be r of R es po nd en t Persons respondent lived with CASE CONTROL University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 42 Figure 4. 10 A distribution graph of the number of other persons case and control live with 4.6.3 SIZE OF RESPONDENTS DWELLING PLACE (N= 240) The number of rooms in the house where respondents live was considered as an indication of the size of the dwelling place.A majority of them, 46.7% (112) stayed in a house with one to two rooms, 40% (96) stayed in three to four rooms houses and 13.3% (32) stayed in houses with more than five rooms. Of the 120 respondents who were non-complaints, 46.7% (56) stayed in house with one to two rooms, 45.8% (55) stayed in a house with three to four rooms and 7.5% (9) in a house with more than five rooms. The number of rooms was however not a significant factor contributing to compliance or non-compliance to TB treatment completion (Table 4.6). 4.6.4 LENGTH OF STAY IN THE CURRENT PLACE (N = 240) The length of stay in the current place was meant to be a measure of some stability for the respondents, thus a factor that may affect their compliance to TB treatment: 7.9% (19) of the respondents had lived in their current houses for less than three months, 21.3% (51) had 0 10 20 30 40 50 60 0 (1-3) (4-6) >7 N um be r of R es po nd en t Number of persons respondent living with CASE CONTROL University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 43 stayed in their houses between three to twelve months and 70.8% (170) had stayed in their houses for more than twelve months, most of the respondents were fairly stable. Of the respondents who had lived in their current place for more than twelve months 46.5% (79) belongs to the non-compliant group. However, the length of stay was not a significant factor contributing to compliance or non-compliance, of TB treatment completion (Table 4.6). Table 4. 6 Default factors in TB treatment for case and control respondents Factors Case N(%) Control N(%) OR(95% CI) PValue Number of people living with 0 (alone) 15 (12.5) 4 (3..3) 1 1 - 3 28 (23.3) 32 (26.7) 0.2(0.07-0.71) 0.011 4 - 6 55 (45.8) 45 (37.5) 0.3(0.11-0.98) 0.047 > 7 22 (18.3) 39 (32.5) 0.1(0.04-0.46) 0.001 Relationship to persons living with 0.004 FAMILY 86 (71.7) 105 (87.5) 1 FRIEND 19 (15.8) 11 (9.2) 2.1(0.95-4.67) 0.066 NONE 15 (12.5) 4 (3.3) 4.5(1.46-14.30) 0.009 Size of household 1 – 2 56 (46.7) 56 (46.7) 1 3 – 5 55 (45.8) 41 (34.2) 1.3(0.77-2.32) 0.249 > 5 9 (7.5) 23 (19.2) 0.3(0.16-092) 0.031 Length of stay in house < 3 months 12 (10) 7 (5.8) 1 3 - 12 months 29 (24.2) 22 (18.3) 0.7(0.25-2.27) 0.635 > 12 months 79 (65.8) 91. (75.8) 0.5(0.19-1.34) 0.173 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 44 4.7 STIGMA AND DISCRIMINATION 4.7.1 DISCLOSED TB STATUS (N =240) The respondents were asked whether they had disclosed their TB status to either a family member or friends: 92.1% (221) of the respondents had disclosed their TB status, 7.9% (19) had not disclosed their status . 73.7% (14) of those who did not disclosed status belong to the non-compliant group. The disclosure of TB status was not associated with any significant difference between the compliant and the non-compliant groups. 4.7.2 REASONS FOR NON-DISCLOSURE (N = 19) Among the respondents interviewed 7.9% (19) did not disclosed their TB status. 84.2% (16) said it was because of fear of being isolated, Out of the 16 who said that it was because of fear of being isolated, 68.7% (11) did not complete their treatment. Three of the non- compliant said they have no one to trust as a reason for non-disclosure. 4.8 DISEASE AND MEDICINE RELATED FACTORS Respondents were asked a number of questions to assess various disease and medicine related factors that could contribute to compliance or non-compliance to TB treatment. 4.8.1 EXPERIENCED SIDE AFFECTS TO TB MEDICINES (N = 240) In relation to experience with side while taking TB treatment. 65.4% (157) of the patients had experienced some side effects while taking medicines while taking medicines, while 34.6% (83)had not.Out of the 157 that experienced side effects, 46.5% (73) did not complete treatment. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 45 4.8.2 TYPE OF SIDE EFFECTS RESPONDENTS EXPERIENCED (N = 157) Those who reported having experienced side effects were asked about the type of side effects they had. Some of the respondents may have had more them one side effects but the questionnaire did not capture these. The commonest side effects reported were headache and dizziness 34.4% (54), followed by 26.1% (41) with skin rashes and 24.2% (38) with painful limbs. Side effects were also common among both compliant and non-compliant groups. Table 4.7 (below) summarizes side effects among the compliant and non-compliant group. Table 4. 7 Summary of the distribution of side effects among the compliant and non- compliant group Side effects Compliant Non-Compliant Headache and Dizziness 30 24 Skin rashes 21 20 Painful Limbs 18 20 Numb foot and hands 6 5 Diarrhoea & vomiting 3 2 Yellow eyes 0 2 Others 6 0 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 46 4.8.3 BEFORE TIME IT TOOK YOU FELT BETTER? (N = 240) Respondents were asked how long it took them to start feeling better on TB treatment. 43.3% (104) of the patients took less than two months to feel better, 49.6% (119) took between two and four months, 8.3% (2) between five to six months, while 6.3% (15) reported they never felt better at all. Out of the 15 who reported that they never felt better at all, 93.3% (14) did not complete treatment. 4.8.4 RESPONDENT’S REASONS FOR NOT COMPLETING TREATMENT (N = 120) The 120 respondents who were non-compliant gave the following reasons: 34.2% (41) inadequate supply of medicines, 30.8% (37) feeling better, 11.7% (14) not feeling better, 8.3% ( 10) lack of family support, 6.7% (8) no reason, 5% (6) stigma and 3.3% (4) side effects as their reasons for not completing TB treatment. Table 4.8 (below) gives a summary of reasons for not completing TB treatment. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 47 Table 4. 8 Reasons given for non-compliance, by cases (defaulters) Reason Number (N) Percentage Inadequate supply of medicines 41 34.2 Feeling better 37 30.8 Not feeling better 14 11.7 Lack of family support 10 8.3 No reason 8 6.7 Stigma 6 5 Side effects 4 3.3 4.8.5 TB DISEASE CLASSIFICATION (N = 240) Of the 240 respondents, 81.3% (195) had smear positive pulmonary TB (PTB SM+), while 18.7% (45) had smear negative pulmonary TB (PTB SM-) . Out of the 195 that had PTB SM+,47.7% (93) were non-compliant, and 60% (27) of those who had PTB SM - were also non-compliant. Among the respondents non belong to the extra pulmonary TB group. There was no significant difference between the two groups with respect to TB classification. 4.8.6 HIV STATUS (N = 240) Of the 240 respondents, 23.3% (56) had an HIV test result positive, while 76.7%(184) had a negative HIV result . 50% (28) of the 56 HIV positive respondent did not complete TB treatment. HIV status among the respondents was equally distributed and was not a significant factor contributing to compliance or non-compliance. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 48 4.8.7 THE TAKING OF OTHER MEDICINCES BY RESPONDENTS (N = 240) Respondents were assessed if they were taking other medicines besides TB treatment, 36.3% (87) were taking other medicines while 63.7% (153) said they were not taking other medicines. Taking other medicines was not significantly associated with compliance or non- compliance . 4.8.8 WHAT OTHER MEDICINES WERE TAKEN? (N = 87) All the 87 respondents who said they were taking , 51.7% (45) were taking Highly Active Anti-Retroviral Therapy (HAART), 5.7% (5) anti-hypertensive's, 23% (20) Psychiatric and 19.5% (17) other medicines. 57.8% (26) of the 45 patients on HAART did not complete TB treatment. Of the 17 patients taking other medicines all belongs to the non-compliant group. 4.8.9 PATIENTS’ OPINION ON WHAT COULD MAKE TB PATIENTS COMPLETE TREATMENT This was an open question to the respondents seeking their opinion in what could make TB patients complete their treatment several responses were given including, 32.1%(77) of the respondent said that family support will help TB patients complete their treatments, 22.1% (53) availability of medicines, 12.9% (31) health talks, 12.5% (30) waiting time, 10% (24) supervision of patients, 5.8% (14) transportation and 4.6% (11) attitude of clinic staff. Table 4.9 (below) gives a summary of the opinion of patients in relation to compliance. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 49 Table 4. 9 Patients opinion with regards to compliance Opinion Number(N) Percentages(%) Family support 77 32.1 Availability of medicines 53 22.1 Health talks 31 12.9 Waiting time 30 12.5 Supervision of patients 24 10 Transportation 14 5.8 Attitude of clinic staff 11 4.6 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 50 Table 4.10. Multivariable model (Binary logistic regression) of factors associated with default from treatment among tuberculosis patients in Western Sierra Leone Odds ratio 95%CI p-value Factors Lower Upper During the time you were taking TB medicines, what would you say was your situation in terms of food availability Always available to take with medicines (RC) 1.00 Available most of the time 6.93 1.52 31.63 0.01 Not always available 2.91 0.24 35.25 0.40 How much time do you usually wait at the TB clinic before being attended? <1hr(RC) 1-2hrs. 0.60 0.25 1.41 0.24 >2hrs 0.09 0.02 0.40 0.00 When you went to pick your medicines at the TB clinic, what would you say about the availability of medicines there? Always available 0.006 0.001 0.052 0.00 Sometimes not available (RC) 1.00 Loss of weight is a symptoms of TB Yes (RC) 1.00 No 0.19 0.03 1.05 0.06 TB treatment should be taken until 6months Yes (RC) 1.00 No 0.007 0.001 0.062 0.00 Who do you live with? Family(RC) 1.00 Friends 5.85 1.81 18.86 0.00 Alone 0.63 0.04 10.85 0.75 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana