1. Introduction

Tuberculosis (TB) is a major cause of illness in low-resource countries. It is among the top 10 causes of death, and has been the leading cause of death from a single infectious agent in recent years [1, 2]. There were an estimated 10.0 million TB cases, and 1.5 million deaths due to TB in 2020 [3]. The 30 high TB burden countries are responsible for the majority (86%) of the estimated incident TB cases [2].

Ethiopia is among the high TB burden countries [3]. TB is a major public health problem and one of the leading infectious diseases in Ethiopia. There were an estimated 140 new incident TB cases per 100,000 population in 2020 and TB mortality rate of 19 per 100,000 populations in 2019 in Ethiopia [4]. Ethiopia accounts for 90,000 (3%) of annually missed TB cases worldwide [2, 5]. A recent national case detection rate of all forms of TB was 76 percent, which is below the target of World Health Organization [3, 6]. One of the contributing factors for low case detection is the delay in diagnosis and treatment of TB. Delay in diagnosis and treatment is commonly divided into three components (patients’ delay, health system’s delay and total delay). While patient delay refers to the delay period from onset of the major TB symptoms to first visit to a medical provider, health system delay encompasses the delay period from first visit to a medical provider to first start of anti-TB treatment. Total delay is defined as the delay period from start of major TB symptoms to first start of anti-TB treatment. Delay in TB diagnosis and treatment remains a major problem of TB control program generally in low and middle income countries [7] particularly in Ethiopia [8, 9]. Patients’ health care seeking delay differs among different regions of Ethiopia [10, 11].

Prompt diagnosis and treatment is crucial for efficient TB control program performance and achieving the End TB targets. The target set for the End TB Strategy for the year 2035 include: 1) reduction of TB mortality by 95%, 2) decreasing TB incidence by 90% (i.e compared to the baseline of 2015), and 3) to certify that no family is suffered with TB related catastrophic costs [12]. The End TB Strategy targets can only be realized if diagnosis, treatment, and preventive services for TB are delivered based on the context of universal health coverage, which implies that all people with TB should be early detected and properly treated [2, 5]. Delayed diagnosis and treatment of TB cases has major role in the transmission of the disease in the community in most high TB burden countries. Early diagnosis and proper treatment of TB will reduce severity of illness, prevent transmission, increase cure rate, and prevent the development of drug-resistant TB [2, 13].

Various studies have been conducted on delays in diagnosis and treatment of TB in different parts of the world [14–21]. A systematic review and meta-analysis conducted in low- and middle-income countries showed that the median total delay ranged from 30 in Zimbabwe and Vietnam to 366.5 days in Afghanistan [7]. Another systematic review and meta-analysis conducted in Ethiopia revealed that the median diagnostic of 45days. The prevalence of diagnostic delay in Ethiopia ranged from 9.57% in Addis Ababa city and 68.84% in Somali region [22]. Various factors that are associated with diagnostic and treatment delay were identified in earlier studies in Ethiopia and elsewhere. Some of these factors included poor knowledge about TB [23], patient first visit to lower level facilities [21], long distance to health facility [19], low level of income [23], being female [19, 24], being illiterate [15, 22], perceived TB stigma [25], rural residence [19, 22], having extra-pulmonary TB [14, 22, 26], having smear negative pulmonary TB [26], self-treatment [21, 23, 26], being HIV negative [22, 23], absence of chest pain and presence of haemoptysis [17] etc.

Although various studies were conducted in Ethiopia and elsewhere to assess TB diagnostic and treatment delay, there is limited study to date that has been conducted to assess the length and associated factors of total delay in diagnosis and treatment of TB in Jimma Zone, Ethiopia. The length of total delay and associated factors may vary according to the local setting, including the socio-demographic and economic condition of the population in the study area [10, 11]. Understanding the contribution of these factors is important to propose targeted interventions to address diagnostic and treatment delay at the local setting. Therefore, the aim of the present study was to determine the length and analyze factors associated with total delay in Jimma Zone, Ethiopia.

2. Materials and methods

2.2. Study population and sampling

A total of 1,161 newly diagnosed TB patients were included from sampled districts’ public health facilities and health posts. Patients whose age was less than 15 years, who could not respond to the interview questions and critically ill patients were excluded from the study. Eight districts and one town administration were selected from the 17 districts and two town administrations by using a simple random sampling method. Subsequently, all TB DOTS sites in the sampled districts and a town administration were covered by the study.

This study is part of a PhD project. It follows a recently published sub-study that compared community- versus facility-based DOTS in a cohort study that enrolled a total of 1,161 study participants [28]. We included all of the 1,161 study participants in the current study. However, in order to address the research objective in the current study, a total sample of 422 calculated using formula for estimating single proportion could have been enough (i.e. considering 50% proportion of delay of more than one month at 95% confidence interval and a margin error of 5%). The inclusion of the 1,161 study participants in the current study is advantageous as it provides more than adequate representative population to address the research questions. The sample size was proportionally allocated to the selected public health facilities based on the previous one year’s (before the study’s start) patient flow. The study participants were consecutively included until the required sample size was achieved [Fig 1].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Schematic presentation of sampling procedure for TB patients, Jimma Zone, 2017.

https://doi.org/10.1371/journal.pone.0281546.g001

3. Results

3.1. Characteristics of the study participants

In this study, we assessed the length of total delay and associated factors among TB patients in Jimma Zone, Ethiopia. Accordingly, a total of 1,161 patients participated in this study: 774(66.7%) under FB-DOTS, 387(33.3%) under CB-DOTS) were included. Of these, 51.2% were male, 65.9% were married, and 39.5% were illiterate. The mean (± SD) age for the study participants was 32.2 (±14.41) years, with a range of 15 to 90 years. Nearly half (46.9%) of the patients were 25–44 years of age. Of the total study participants, 63.7% were farmers by occupation, 51.7% were rural residents, and 48.8% were smear-positive pulmonary TB patients (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Socio-demographic and clinical characteristics of the study participants in Jimma Zone, 2017 (N = 1,161).

https://doi.org/10.1371/journal.pone.0281546.t001

3.2. Study districts and time to reach the nearest medical provider

A majority (18.9%) of the study participants were from goma district. The mean (± SD) time taken to reach the nearest medical provider was 32.86 (±28.52) minutes, ranging from two to 180 minutes (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Districts and time to reach the nearest medical provider of the study participants at Jimma Zone, 2017 (N = 1,161).

https://doi.org/10.1371/journal.pone.0281546.t002

3.3. Patients’ first visit to medical providers

A total of 708 (60.98%) patients visited other medical providers (health posts and private clinics) before visiting the current health facility where they got their TB diagnosis and initiated anti-TB treatment. Of these, a majority (169) of the patients who first visited private clinic were not delayed. On the contrary, most (45) of the patients who visited health posts were delayed (Fig 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Medical provider first visited for the symptoms with patient delay status at Jimma Zone, 2017.

https://doi.org/10.1371/journal.pone.0281546.g002

3.4. Patient delay and total delay

The median patient delay (time from onset of symptoms until first visit to a medical provider) was 30 days [interquartile rang (IQR): 17.5, 60 days]. Whereas, the median total delay was 35 days (IQR: 25, 67 days). The total delay ranged from 4 to 732 days (Table 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Patient, health system and total delays for the study participants in Jimma Zone, 2017.

https://doi.org/10.1371/journal.pone.0281546.t003

3.5. Determinants of total delay

The median total delay was 35(IQR 25, 67 days) days, and for more than half (586) of the patients, the total delay was > 35 days. Variables such as occupation, source of household income, being a smoker, alcohol use, khat chewing, HIV status, TB classification, chest pain, and weight loss were removed from the final model even if these were statistically significant during the bivariate analysis. In the multivariate analysis, patients who had poor knowledge about TB were about four times more likely to be delayed than patients who had good knowledge about TB (AOR = 3.92, 95% CI: 2.65, 5.80). Those patients who had swelling or wound in the neck region were three times more likely to be delayed than their counterpart (AOR = 3.02, 95% CI: 1.62, 5.62). Patients who attended college or university were 72% less likely to be delayed than those patients who did not read and write (AOR = 0.28, 95% CI: 0.10, 0.81). Female patients were about 1.5 times more likely to be delayed than their counterparts (AOR = 1.46, 95% CI: 1.00, 2.14). Patients who had a monthly household income of 1001–2500 ETB had increased total delay compared to patients who earned > 3500 ETB (AOR = 15.75, 95% CI: 2.92, 84.91). Those patients who traveled 30 or less minutes to reach the nearest medical provider were 53% less likely to be delayed than those patients who traveled more than 60 minutes (AOR = 0.47, 95% CI: 0.23, 0.94). Patients under CB-DOTS were about 1.5 times more likely to be delayed than their counterparts (AOR = 1.53, 95% CI: 1.02, 2.29). Patients who had no limitation of their day-to-day activity were less likely to be delayed than those patients who had high limitation of their day-to-day activity (AOR = 0.09, 95% CI: 0.01, 0.89) (Table 4).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Determinents of total delay among the study partcipants at Jimma Zone, 2017.

https://doi.org/10.1371/journal.pone.0281546.t004

4. Discussion

In this study, we assessed the length and associated factors of total delay among TB patients. A systematic review and meta-analysis conducted in low- and middle-income countries reported that the median total delay ranged from 30 to 366.5 days [7]. We observed a median total delay of 35 days which is similar to the study done in Addis Ababa, Ethiopia [35], but lower than the findings reported from earlier studies conducted in different Regions of Ethiopia [23, 25, 36–38]. The shorter total delay observed in the present study may be related to several reasons: these may include improved access to TB diagnostic and treatment facilities through decentralization of health services to the community, and health extension workers’ contribution in identifying suspected TB cases and referring them to health centers for smear microscopy test [39].

Even though the observed total delay in our study was relatively lower compared to earlier study findings in Ethiopia, it is still long delay given the need for immediate diagnosis and treatment initiation for TB patients. In a recent qualitative study conducted in the study area, different barriers were identified that may have implications for increased total delay. Among these were shortage of resources including basic infrastructure and limited TB diagnostic services; and delays in the diagnostic process at the health facilities [40]. Solving these barriers through targeted interventions is crucial to reduce the observed relatively long total delay. Long delays in diagnosis and treatment initiation increase the severity and complications of disease that may result in unfavorable treatment outcome [37, 41]. It also increases the risk of developing drug resistant TB which leads to treatment failure, high mortality rates, and high transmission rate of drug resistant TB [42, 43]. These findings emphasize the importance of early diagnosis and prompt treatment of TB. Conversely, the observed median total delay in the current study is higher than the study findings in other parts of Ethiopia that reported 23 days and 33 days [8, 9]. This discrepancy might be because of differences in the study settings i.e. socio-economic status, study population, access to TB diagnostic facilities, and utilization of health care services among the population.

The analysis of socio-demographic, economic, and clinical factors revealed significant associations with increased total delay. In line with our study, low level of education [7, 25, 36, 38], long distance [19, 36], and poor knowledge about TB [23, 36, 37] were identified as predictors of total delay in former studies in Ethiopia and elsewhere. Our findings underscored that educational level is an important determinant of total delay. A former study in Ethiopia revealed that individuals with a higher level of education had better knowledge about TB compared to those who did not read and write [44]. Our study showed that the educational level of the patients may not be directly associated with the delay but do significantly affect the knowledge level of the study participants. A study from the Gambia showed an association of educational level with knowledge, attitude and health-seeking behavior regarding TB [45]. Another study from Malaysia reported that TB education program at school was an effective intervention for improvement in the mean score of knowledge, preventive practice, and perceived stigma about TB [46]. In addition, a study conducted in Ethiopia indicated that there was association between educational level and knowledge about TB in which having good knowledge of TB led to a more positive attitude and better practice in relation to TB prevention and control [47]. This was supported by findings from a qualitative meta-synthesis conducted in Nigeria which reported that low education level and poor knowledge about TB were major barriers to TB diagnosis and treatment [20]. Compared to the uneducated persons, individuals with better educational level are more likely to have access to better economic situation which consecutively improves their access to health information and care [8, 44]. Moreover, the level of their knowledge about TB influence their perceptions about TB and change their health care-seeking behaviour [8, 44].

We observed a strong association between having swelling or wound in the neck region and long total delay. Cervical TB lymphadenitis is one of the most common manifestations of extra-pulmonary TB which contributes to diagnostic delay. The common sign and symptom is a chronic, painless swelling in the neck region (no other remarkable symptom). Thus, the diagnosis of this type of TB requires a high index of suspicion and use of different diagnostic methods. Practically, it is not possible to apply all diagnostic procedures for all symptomatic patients. In this regard, ruling out for other causes of neck mass could contribute to diagnostic delay. Because of the absence of culture and pathology services in most health centers in Ethiopia including the study area, it is difficult to early diagnose such type of TB [48, 49]. A study from Zanzibar showed that from all patients with long total delay (> 6 months), a majority (90%) were patients with TB lymphadenitis [50]. Extra-pulmonary TB cases are seldom infectious as they are considered to have less contribution in TB transmission. However, delay in starting treatment for the patients may result in disseminated TB and increased mortality [49, 50]. As peripheral health care facilities in the study area do not have the capacity to diagnose extra-pulmonary TB cases, it is important to develop diagnostic algorithms for timely referral of presumptive extra-pulmonary TB patients. Therefore, training is essential for the health extension workers so that they can screen TB lymphadenitis suspects in the community for prompt referral to the next level of health care for better diagnosis and treatment [51].

Our study revealed that women were more likely to have long total delay compared to men. This finding is consistent with former studies conducted in Ethiopia and elsewhere [36, 52, 53]. Economic status, cultural beliefs, and perceived stigma are barriers of early care seeking for most female patients [54]. In terms of cultural beliefs, women in Ethiopia give priority for their family than themselves, have main responsibility for children care, have less decision making power, and worries about preserving modesty might affect their health care seeking behavior [55]. Previous study revealed that women were more likely to suffer from TB stigma compared to men [56]. Women in high burden countries like Ethiopia experienced long delays in TB diagnosis and treatment because of challenges related to TB services [57]. Most women faced problems in accessing TB service because of resource limitations, power imbalances, and poor knowledge about TB [57]. Women who have less access to quality healthcare, and delayed to seek formal health care were prone to long delays [54]. This may partly be related to the fact that women have the main responsibility for domestic tasks and care for their family members, especially children and elders. Women often lack economic independence, have less time and are less empowered compared to men in regards to care for themselves [36]. Therefore, it is important to recognize sex differences in individual’s TB care-seeking. Targeted interventions that enhance women empowerment are required to improve poor health outcomes among women. Socio-economic support interventions could change their health seeking behaviour, improve adherence and treatment outcome for TB patients who are women [58, 59]. In addition, gender-based health education and behavioural change communication (BCC) interventions focusing on early health seeking counselling are required to reduce the burden of TB among women. BCC strategies are efficient, effective, sustainable and acceptable because messages can tailored to gender- or age-based target groups [60, 61].

Monthly household income was another determinant variable that was significantly associated with total delay. This finding is in line with previous study results from Ethiopia [8, 62, 63] and Pakistan [18]. This could be due to the fact that individuals who did not have adequate monthly income could not be able to seek health care early because of several reasons such as unaffordable medical costs and transport expenses [8, 63]. Therefore, they are likely to be prone to delayed diagnosis and treatment start. In resource limited settings like Ethiopia where there are several traditional practices and low access to quality health service, patients commonly seek health care from informal medical providers, as a result the patients might get traditional treatment that could delay them from timely seeking medical care from formal medical providers. TB diagnosis and treatment services in Ethiopia is given for free, money is mostly needed for covering costs for transportation, accommodation and food for the patient. Individuals who have low level of income may need to work long hours, as a result they may not have time to early seek health care when they get ill. Patients with inadequate income are more likely to visit formal-health care provider only when they become critically ill [8, 63]. A systematic review conducted in high TB burden settings reported that livelihood, work, and family were given priority which led them to have a long delay in health care-seeking [57]. This finding shows the importance of improving the socio-economic condition of the population for effective TB control. As per the Sustainable Development Goals (SDG), the world is committed to end poverty by 2030 [64, 65]. According to a recent World Bank report, the poverty rate in Ethiopia fell from 44 percent in 2000 to 21 percent in 2018 [66, 67]. If this trend continues, the contribution of poverty to TB disease and transmission may be significantly reduced in Ethiopia including the study area.

Time taken to reach the nearest medical provider was associated with total delay. Long distance traveling to reach the nearest health facility was also reported in previous studies conducted in Ethiopia and other countries as barrier to individuals’ health-seeking behavior which leads to lengthening of patient’s delay [17, 68–72]. A recent qualitative study from the study area identified that long distance traveling to get TB diagnostic services was a barrier for most TB patients from the rural settings [40]. This could also be related with patients who were under FB-DOTS travel longer distance and more delayed than those patients under FB-DOTS.

Our study showed that patients under CB-DOTS were more likely to be delayed than those patients under FB-DOTS. This could be due to the fact that patients under CB-DOTS were rural residents and most of patients under FB-DOTS were urban residents. Rural residence was reported as independent predictor of delay in a systematic review conducted in Ethiopia [11]. One of the reasons might be that rural residents had poor access to health information and TB diagnostic and treatment facilities compared to urban residents. This might be due to the scarcity of formal health care providers in rural area of Ethiopia. Thus, patients from rural areas might take long chains of care-seeking through informal health care providers [57, 73]. A previous study from Southwest Ethiopia revealed that proportion of households sought care from formal health care providers was lower among rural compared to urban households [74]. Moreover, rural residents lack regular health information about the disease and do not seek health care early compared to the urban residents [10, 11]. In addition, rural residency makes it hard to travel to diagnostic and treatment facilities due to absence of road or distance (as long distance travelling discourages initiation of tuberculosis treatment) [10, 11].

This study has strengths and limitations. The strength of the study is that we included a relatively larger sample size compared to sample sizes used in other similar studies in Ethiopia. Thus, it provides more than adequate representative population to address the research questions. With regards to potential limitations, the study was only carried out in government health facilities; hence the findings cannot be generalized to all TB patients in the study area. There are also several private health care facilities and non-formal health care providers whereby TB patients may seek health care for their symptoms. Moreover, the present study focused on adult TB patients aged > 15 years who were treated at public health facilities; therefore, the results among other age groups and in other similar settings during the study period might be different. In addition, the reported duration of symptoms and first visit for treatment seeking is based on patients’ ability to recall and interpretation of their symptoms. Since patients might not exactly remember the exact date of start of their symptoms and first visit to a medical provider, it is subject to a recall bias. However, we have tried to reduce the recall bias by using different techniques such as using local calendars linking to major religious and national days to explain their perceived date of onset of TB symptoms and first visit to a medical provider.

5. Conclusion

The present study showed long total delay (the median total delay of 35 days) in diagnosis and treatment start for TB patients in the study area. The study also revealed important factors associated with total delay. Poor knowledge of TB, swelling or wound around the neck region, being women and low level of education were identified as factors associated with total delay. Several interventions are required to reduce the observed length of total delay. Efficient implementation strategies for early case detection and treatment initiation should be practiced to shorten total delay. Among others, well-crafted IEC and BCC strategies on TB need to be designed to increase awareness and health service utilization among the population in the study area. Swelling or wound around the neck region in a TB patient may be linked to TB lymphadenitis or disseminated TB. The health care facilities at peripheral level in the study area do not have the capacity to diagnose TB lymphadenitis cases. Thus, it is important to develop diagnostic algorithms for timely detection and referral of presumptive TB lymphadenitis patients to the nearest medical providers with the capacity to diagnose TB lymphadenitis. Then, the delay period among this group of patients is shortened and prompt treatment is initiated on time.

Supporting information

Acknowledgments

Our gratitude goes to the University of Oslo, Institute of Health and Society and Jimma University for supporting us to undertake this study. We would like to thank the Oromia Health Bureau, the Jimma Zone and Jimma Town health offices and all the study participants for providing us with the necessary information.