Background
Tuberculosis (TB) is a major public health challenge, with an estimated 8.8 million new cases and 1.7 million deaths in 2003 [
1]. 27% of these cases and 31% of the deaths occur in Africa, where case rates continue to rise. The situation in Africa has been attributed to the HIV epidemic and poorly organized TB control programs [
1,
2]. Treatment of TB requires access to appropriate health care, but patients may find it difficult to adhere to the intended treatment even if such services are available. Studies from both low- and high-income countries show that between 20% and 50% of patients with TB do not complete treatment regimes [
3,
4]. Incomplete treatment may result in prolonged infectiousness, drug resistance, relapse, and death [
5]. Improvement in treatment outcomes requires a better understanding of the barriers and enablers patients experience during TB treatment. Quantitative and qualitative studies from countries in Asia and Africa point to barriers related to availability and accessibility [
6‐
8], and direct and indirect costs related to treatment [
9‐
12]. Inconvenient routines in health care systems [
13‐
15] and interaction with health personnel may also act as barriers [
13,
16‐
18].
In Ethiopia, where there is an increasing incidence of new infectious TB cases [
19], quantitative studies from rural areas found that between 6.7% and 20% of patients interrupted treatment and that long distances to the health facilities, poor awareness about the disease and treatment length, side effects and lack of family support are important factors [
20‐
23]. An early case control study from Addis Ababa found that "social problems" and "feeling of improvement" were major causes of treatment interruption [
24]. These studies found that most patients interrupted treatment in the third or fourth month.
Few studies have used qualitative methods to explore patients' experiences during different phases of treatment, and to the authors' knowledge the studies on diagnostic delay and treatment interruption in Ethiopia have not been explored further in qualitative research.
WHO has emphasized that an important research area is identifying time points in the treatment that are of importance for different types of adherence strategy [
25]. The aim of the present study is to explore enablers and barriers in the management of TB treatment during the first five months of treatment in Addis Ababa, Ethiopia.
Results
We found that loss of employment or the possibility to work led to a chain of interrelated barriers for most TB patients. Daily treatment was time consuming and physically demanding, and rigid routines at health clinics reinforced many of the emerging problems. Patients with limited access to financial or practical help from relatives or friends experienced that the total costs of attending treatment exceeded their available resources. This was a barrier to adherence already during early stages of treatment. A large group of patients still managed to continue treatment, mainly because relatives or community members provided food, encouragement and sometimes money for transport. Lack of income over time, combined with daily accumulating costs and other struggles, made patients vulnerable to interruption during later stages of treatment. Patients who were poor due to illness or slow progression, and who did not manage to restore their health and social status, were particularly vulnerable to non-adherence. Such patients lost access to essential financial and practical support over time, often because relatives and friends were financially and socially exhausted by supporting them. These findings are grouped into main themes and examined in further detail below.
Loss of income
The majority of patients in the study experienced loss of employment or the opportunity to work as one of their main problems during treatment. Patients reported losing their job when their TB diagnosis was known, or because they were too ill to continue working, or were unable to find daily work because of the time consuming treatment arrangements. Three patients were housewives, but the household finances were still affected because their husbands lost income and risked their jobs escorting them to the clinic. One of the husbands explained:
"I tried to explain for them that I was the only one working in our family, so I couldn't go every morning ... because in that way I would lose my job and me and my wife would die of hunger. I work in the private sector and it is difficult to get permission every day." (Participant 24, prospective group.)
Patients who worked in the private sector and "daily labourers" (a person who is not employed on a permanent basis, but who may meet at a regular point every day to compete with others to be hired for the day) were most affected. Also health personnel identified these groups as particularly vulnerable:
"People who are daily labourers or work in the private sector they can't get any kind of [sick] leave and they face more problems. Many of them interrupt treatment because they don't want to lose their jobs. They rather live for a while, with money and their jobs, to eat and then die." (Participant 27, male nurse.)
Being ill and without access to any health care benefits often resulted in a loss of income from one day to the next. Patients' situations were aggravated because loss of income coincided with additional expenses. More than half of all the patients had been treated at private clinics before entering conventional TB treatment, paid for from savings or money borrowed from relatives or community members. Daily attendance at a clinic caused high transportation costs, and those that could not afford transportation incurred opportunity costs due to extensive time use. In addition, most patients believed that they had to eat expensive protein food like meat, milk and eggs to get cured. A daily labourer offered an illustration of how the struggle to obtain "good food" increased his already pronounced personal crisis caused by loss of income:
"How can I get this food [good food], it is impossible to get this. I don't have a job. [...] "I'm even considering committing suicide. [...] I have lost all hope ... totally. I am the one that support me and now I can't support myself. I can't work, and then it is hard for me to survive." (Participant 10, male patient, prospective group.)
Loss of income combined with additional costs made TB patients' particularly vulnerable during the first period of treatment (the intensive phase). Two patients in the prospective group continuously considered interrupting treatment due to a lack of money for food and transportation to the clinic.
Hunger
Impoverishment due to loss of income caused unpredictability and emotional stress in relation to daily access to food. Many participants described feeling chronically hungry, suffering the slow starvation of people who eat a little bit every day but whose hunger is never fully satisfied. A young male patient, who lived with his single, TB inflicted mother, was too ill to continue working as a daily labourer. He related how lack of food made him lose his motivation to follow the treatment:
"Q. During the last weeks, have you ever considered not to go to the health centre?"
"Yes, I thought about that many times ... on and off. Especially when I lack food I feel like not coming. [...] My only motivation is to get food ... food ... it is food that is motivating me [...] to get food, any kinds of food ... at the right time ... and to have a person that can help you get that food." (Participant 8, male patient, prospective group.)
Chronic hunger induced a feeling of hopelessness among patients. Irregular and insecure access to food over time made many lose their hope of recovering and avoiding death.
Taking medicines on an empty stomach
Gastritis was the most common side effect, and was reported by nearly all intensive-phase patients. The symptoms were described as a burning or gnawing pain in the abdomen. A few patients explained that they pretended to swallow the medicines, saving them until they could take them with food or milk. A female focus group participant explained:
"I knew that I would feel the pain if I swallowed the tablets immediately [at the clinic]. Therefore I pretended swallowing the tablets. I took the tablets, hiding them. Then I swallowed the tablets later ... with milk ... at home." (Female focus group participant, focus group one.)
Patients attributed side effects, such as gastritis, nausea, and vomiting, to taking strong medicines on an empty stomach. These side effects had a large psychological impact: Since most patients considered access to food, and particularly food with a high content of protein, as extremely important to healing, symptoms as gastritis served as a continuous reminder of their poverty and what they considered to be poor healing conditions.
Physical demands
Most patients complained that the first two months of treatment were physically exhausting. This was particularly the case for the many patients who were at advanced stages of the disease at the time they started treatment. Patients attending treatment at Kebele 16 Health Post had shorter walking distances than those attending the other clinics. This was a tremendous advantage for the many who could not afford transportation. Most patients from the other two clinics reported using one to two hours each way. A patient with advanced symptoms, and who was also suffering from various side effects, described her daily walk to the clinic:
"Q. Do you always walk or do you take any kind of transport?"
P. "I walk, I don't have any money for transport."
"Q. Is it hard?"
P. "Yes, it is a bit far. I use almost two hours and I vomit on the way." (Participant 5, female patient, prospective group.)
Once at the clinic, patients often waited an hour or more to get their medication. This delay was mainly caused by intensive-phase patients being told to attend at the same time in the morning. This time-consuming system was described as humiliating by some of the patients.
Rigid routines and health staff attitudes
The three clinics in the study practised DOT without taking into consideration that many patients were not physically able to come to the clinic. Many patients told how they had begged to be hospitalized, but due to a lack of hospital beds this option was out of the question for many. One patient died on the 59th day of treatment, having been refused medical attendance before the 60th day, in accordance with clinic routines. Her husband described how they had struggled to make health personnel understand that she was far too ill to walk to the clinic:
"We went to the clinic to ask if the nurse could give her the medicines, at least once in seven days, telling him that our house was far away [...] and explaining them that she was too weak to go to the clinic every day ... she couldn't walk at that time. The nurse didn't understand, rather he threatened her saying he would demand her to come and collect the medicines no matter how ill she was." (Participant 24, prospective group.)
In a few cases, relatives were allowed to collect the medicines on behalf of very ill patients. This, however, put the relatives' jobs in jeopardy, as they had to queue at the clinic every day. In most cases, the patients were escorted to the clinic by relatives. This incurred transportation and opportunity costs for two persons, often within the same household. The clinics allowed no exceptions for family- or cultural events, and were highly inflexible as regards patients trying to combine daily treatment with work-related activities. Only one patient in the prospective group managed to hold on to his job. He explained the repeated difficulties he faced trying to balance the demands of work with the treatment arrangements:
"Once I came late and I met the bad nurse ... and he was angry about it. He told me to sit and wait for him until he got back from town, just to give me my medicines. After he warned me not to come late anymore and then he gave me the medicines. He said, 'next time you will be given the injections even later and you will be very late for your work'." (Participant 6, male patient, prospective group.)
Some nurses were more flexible than others, but there were examples from all three clinics of patients who were threatened, humiliated or treated angrily by staff for not adhering to the implicit rules of the system. At one of the clinics, two patients reported that they had been denied access after a period of interruption.
Dynamics of social support
The support of family and community members was extremely important during the intensive phase of treatment, because this to a large extent compensated for loss of income. Even though TB caused fear and stigmatization, people shared resources in a time of crisis, and reciprocal arrangements provided most TB patients with some food. Many received physical support in walking to the clinic, and some received money for transportation. A male patient who lost his job as a car washer explained how the contributions of relatives and community members helped him:
"Even if I can't get good food I try to get as much food as possible. People who visit me bring me food ... neighbours and relatives." (Participant 9, male patient, prospective group.)
Many patients experienced changes in the level of support as the treatment program progressed. A male patient, who lost his job as a temporary teacher, moved to live with relatives in a rural area after completing the intensive phase. His friends and grandmother no longer had the means to help him:
"She [grandmother] couldn't afford to help me anymore. She couldn't help me because of her shortage. They are all fed up [...] I have moved, walked from place to place ... and there is no food. In the rural area, even if my father is poor, I at least managed to get milk from the neighbourhood." (Participant 1, male patient, prospective group.)
Many patients faced difficult situations after completing the intensive phase of treatment, as the levels of support dropped. Both patients and health personnel reported that the causes of the changes in support levels were closely related to practical and symbolic changes in treatment at this stage.
Crisis precipitated due to completion of the intensive phase
Based on a synthesis of the various sources of data we found completion of the first phase of treatment to precipitate changes, often critically affecting existing support mechanisms: First, patients were expected to stop going for treatment on a daily basis, and were not expected to need money for transport or help with walking to the clinics. These changes reduced the attention focused on their difficulties, and their needs therefore became less visible. Second, it was expected that the health status of patients who had finished the streptomycin injections, which were considered the "main" medicine, should improve significantly. Third, physical improvement was associated with the expectation that patients would start to work again. Fourth, the foods provided from the community were expected to make a physical difference in patients. There is a strong association between TB and HIV/AIDS, and patients with a "curable disease" are expected to gain weight and strength within a certain period of time. Patients with prolonged disease (intermittent symptoms or no weight gain) may be treated as having an incurable disease (AIDS or a "chronic" type of TB), a condition that carries additional stigma.
Six out of 10 patients in the prospective group, all still financially dependent on others, experienced increasing problems in mobilizing enough help in the period between the second and fifth months. Some gradually received less food, others were told to leave the house because they were no longer able to contribute with the rent. The experience of one patient, who still displayed symptoms as skinniness and weakness during the continuation phase, is illuminating. He reported how his relationship with his brother and sister-in-law became increasingly tense because he could no longer pay his share of the food and rent. He was told to leave in his fifth month of treatment:
"People will get bored of you. Earlier they supported me when they had money, or with some food, but they don't do that much more. I can't ask for help all the time. (Crying) It is at home ... they have asked me to leave their house now." (Participant 10, male patient, prospective group.)
Relatives and community members may start to interpret the slow recuperation process negatively, as indicating eventual death, or they may assume that the patient will not mange to re-establish his or her income. The basis of reciprocity is taken away both when a patient become totally dependent either permanently or for a prolonged period of time, and when people think that the patient is dying. When health personnel were asked about their experiences related to treatment interruption, they all mentioned patients leaving to live with relatives in rural areas as one of the main causes. This was seen as a consequence of patient-relative relationships being financially or emotionally exhausted in the later stages of treatment.
Interplay of factors
Attending TB treatment brought with it various struggles to meet the physical, psychological and financial costs of treatment. Interviews with patients who had interrupted treatment also demonstrated the interplay of such factors. Five patients interrupted during the intensive phase of treatment, two of them with only a few days left. Six patients interrupted between the third and sixth months of the continuation phase. Only one patient pointed to a single factor as the cause of treatment interruption. This patient was denied a sufficient ration of medicine to enable her to travel to the place where her son just had been murdered. The causes of interruption were different, but interrelated. They often operated over time and were most often related to financial constraints. Three examples that illustrate identifiable patterns in treatment interruption are presented below.
Case one
A single mother, a working immigrant from a rural area, used to beg outside one particular church before and after morning mass. She had a small but regular income from regular church visitors. Due to DOT, which takes up the morning hours, she couldn't do that anymore, and her only source of income was dramatically reduced. As a result, she was unable to pay for the room she rented, and she and her baby ended up on the street. She turned to the health personnel, telling them that she was starving and that she needed to take the medicines with her to her mother in the countryside. Her request was denied, and she decided to leave without the medicines after two weeks of treatment:
"I didn't intend to interrupt the medicines at that time, but I was too poor to keep on taking the medication. I was told to eat eggs, meat and to drink milk, but the truth was ... I didn't even have injera [Ethiopian pancake]. I went to the countryside because I was forced to go there." (Participant 19, female patient, retrospective group.)
This case illustrates how poor patients without rights to paid sick leave or access to financial support from family and friends experience acute and unbearable financial crises during early stages of treatment.
Case two
A male TB patient interrupted treatment shortly before completing the intensive phase. During treatment, he made intensive efforts to find daily work, but his physical condition and the time-consuming system made this difficult. He often had no money for transport, and he felt weak and tired from having to walk four kilometres twice a day. Sometimes he was late for treatment, which made the health personnel angry. After about a month and a half, he experienced that the friends who had helped him the most became more reluctant to support him with money. At the same time his relatives, all living in a rural area, invited him to a funeral. He asked the health personnel to be allowed to attend, but this request was denied. He decided to leave, interrupting treatment. When asked why he interrupted, he gave several reasons:
"P. I didn't finish the treatment because I had to work to get food and I was told to eat anything possible for the medication and for the disease ... otherwise the medicines can't work ... and to get food I had to work. The other thing was that I didn't have money for transport to come to the clinic and I couldn't walk always because I would be too late by the time I arrived at the clinic. [...] I felt weak by walking to the clinic everyday, and the insult of health personnel discouraged me.
Q. Would you say this whole situation made you go to your relatives?
P. I went there because I was obliged to go there. (Pause) But I was not motivated to come on foot to the hospital anymore." (Participant 13, male patient, retrospective group.)
This case illustrates the situation where a patient ends up in a vicious circle, where the demands of treatment become more and more difficult to handle. A funeral or a sick relative then represents a breaking point; a legitimate way out of daily and accumulating struggles.
Case three
A young girl interrupted treatment after four and a half months. At the time she started the medication, she was too ill to walk to the clinic, and her mother collected the medicines every day on her behalf. The time the mother spent on daily attendance and nursing her daughter made it impossible for her to work, and both mother and daughter became totally dependent on food provided by their neighbours. After two months of daily treatment, the community members expected them to manage on their own again. However, re-establishing herself as a beggar and laundry washer proved difficult for the mother, and as she was the only breadwinner, she became unable to feed all her children. Desperation and hunger made the sick girl break with the established social norms of the community by begging in her own neighbourhood. This led to unbearable fights with her mother:
"I started with the treatment, but after a while I felt so hungry and the shortage of food at home ... and because I felt like mentally disturbed taking the medication ... made it impossible for me to continue. On top of that my mother treated me so badly at home. To overcome the shortage of food at home I started to beg food and money from the neighbourhood. This begging thing made my mother really mad. She said I was humiliating her. We had some really big fights and after that I escaped from home." (Participant 21, female patient, retrospective group.)
This case illustrates how prolonged illness may affect relations with relatives and community members. Weeks or months of involuntary dependency force patients into humiliating situations, while relatives and friends may suffer from social or financial exhaustion.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
MS initiated the research, wrote the research proposal, conducted the research, analyzed the data and wrote this paper. JCF contributed to the conception and design of the study and the analysis and interpretation of the data. He also contributed to the writing of this paper by critically revising it. GB contributed to the conception and design of the study. He also contributed to the writing of this paper by critically revising it. All authors read and approved the final manuscript.