Tuberculosis (TB) is a community-acquired disease whose infectious agent was before the COVID-19 pandemic the leading cause of death from a single infectious agent worldwide [1]. Detection rates are still under 90% targeted by World Health Organization (WHO) for TB elimination [2]. In 2020, the number of newly diagnosed cases was 5.8 million, far from the 10 million estimated cases worldwide meaning a decline of 18% attributable to the effect of the pandemic of COVID-19 [1]. Longer diagnosis delays contribute to the spread of TB infection. Several strategies for TB detection and treatment have been implemented in countries for several years such as the Directed Observed Treatment Strategy (DOTS), still, many TB cases continue to escape the health system, leading either to irreversible complications or death. In 2020, 1.5 million people died of tuberculosis, including 214,000 cases of TB/HIV co-infection [1]. Lungs are the preferred organ for TB bacteria localization and spread, the delay in diagnosis is a factor of poor prognosis. In the Democratic Republic of Congo (DRC), a study showed that the risk of death was 5 times higher in smear-positive pulmonary tuberculosis than in other forms of tuberculosis [3]. In West Africa, a study conducted in Guinea Conakry on TB diagnosis delay reported a median delay of 11 weeks and a half (54%) of the patients had used non-conventional medicine [4]. A systematic review and meta-analysis reported that in sub-Saharan Africa the time from symptom onset to the diagnosis of TB varied between countries, ranging from 4 days in Sudan to 63 days in Ethiopia with a median of 28 days [5]. In Kigali, a study found a total diagnosis delay of 57 days and identified that the trial of antibiotics by the healthcare system increases the risk by 3 folds of delaying TB diagnosis [6]. A literature review on common TB diagnosis delay risk factors has reported several modifiable gaps such as low knowledge of TB symptoms, old age, social belief, repeated visits in health centers having the same level, and initial consultation of traditional therapy. In developed countries, the diagnosis of TB is also delayed; in the United Kingdom, a delay of 49 days was reported in London [7]. In Portugal, a median total diagnosis delay of 62 days was observed, and there was an increase in the total diagnosis duration between 2008 and 2017 [8]. In countries with a high burden of TB, longer delays can significantly contribute to the spread of the disease in the community. In India, a total diagnosis delay of 130 days was reported; smoking and being part of the low-income group were found to be risk factors for delayed diagnosis [9]. In Mali, there is a lack of data neither on TB diagnosis delays nor on risk factors and reasons for longer delays, thus, this study aimed to identify risk factors for pulmonary TB diagnosis delays in Mali.

Study Design and Setting: A cross-sectional analytic study was conducted from January 2019 to December 2020 at the service of pneumology of the University Hospital of Point-G in Bamako, Mali.

Study Population: Patients diagnosed with positive microscopy pulmonary TB and treatment initiated at one of the six referral health centers (RHC) of Bamako.

Inclusion criteria: Patients who were aged 18 years or more, have been diagnosed with pulmonary TB by sputum microscopy and/or GeneXpert MTB/RIF^®, and have verbally consented to participate in the study were interviewed.

Exclusion criteria: Patients with a diagnosis of tuberculosis who were not able to participate because of the level of illness and those who did not consent to participate.

Procedures and data collection: Patients with sputum positive for Acid-Fast Bacilli (AFB) by microscopy and/or the presence of the genome of Mycobacterium tuberculosis (MTB) by GeneXpert MTB/RIF^® are routinely referred for anti-tuberculosis treatment initiation. These patients were approached for a brief description of the study, verbal consent, and data collection. Information was collected during a one-on-one interview with the participant using a pre-established questionnaire (appendix). Data collected included socio-demographic variables including age, gender, occupation, marital status, level of education; diseases symptoms; smoking status, chest x-ray results; facilities visited, and the average cost of the expenses before the test that confirmed TB diagnosis.

Definitions: Patient delay was defined as the duration between the onset of TB symptoms (cough plus one of the following symptoms night fevers and/or sweat, weight loss) and the first medical consultation (healthcare visit). The medical (health) delay, the duration between the first consultation at a medical center and the TB diagnosis. The diagnostic (total) delay, the period between symptoms onset, and the diagnostic test result. The diagnosis delay was the sum of patient delay and medical delay.

Data analysis: Data were de-identified, entered in Excel 2010, cleaned, and analyzed with SPSS 25.0 for Windows. After sample description of sample characteristics, ANOVA test was performed to compare means, Chi² test was used to determine associations, a logistic regression analysis was performed including variables that had an association with diagnosis (total) delay of more than 100 days with a p-value ≤ 0.25. Any difference was seen with a probability of p < 0.05 was considered statistically significant.

Ethical Considerations: An agreement to conduct the study was provided by the Director of the University Teaching Hospital of Point G before the study was conducted. A Verbal consent was obtained from each patient before data collection. Data were anonymized before analysis and results are presented in aggregates format.

A total of 266 patients were included in this study from January 2019 to December 2020. Males were predominant with 80.8%. The age group [26 - 50] years represented 75.2% with a mean age of 40.5 ± 11.9 years, half (50.4%) of the patients had a primary level of education, 74.1% were residents in Bamako and 25.9% came from different regions of Mali. Two-thirds of patients (65.4%) had a monthly income between 100,000 and 200,000 CFA around 180 - 360 US Dollars. Smoking was found in 42.1% of whom 81.2% had a tobacco consumption between 0 and 20 packs/year. Clinical symptoms found in addition to cough were dyspnea (63.5%), chest pain (76.3%), and hemoptysis (12.8%). The body mass index (BMI) was below 18.5 kg/m² in 79.7%. The WHO performance status (PS) score was either 1 or 2 in 66.2%. Chest X-ray showed bilateral infiltrates in 50.4% and excavation was present in 17.7% (Table 1). Consultation of traditional health providers or self-medication was found in 44.4%. Before TB diagnosis 53.0% consulted in 1 to 2 health facilities and 47.0% in 3 or more clinics. Among health structures visited, 60.9% were communal health district (CSCOM), pharmacy or dispensary (44.0%), and private clinics (39.9%). Different reasons were provided by the patients to justify the time between symptoms onset and the diagnosis, test non requested by a health worker (medical reason) was the most found (79.7%), personal reasons (38.3%), and financial reasons (33.5%), social reasons (30.8%) and professional reasons (30.5%). The estimated cost of patients’ expenses before diagnosis was between 50,000 and 100,000 CFA in 35.7% and spent more than 100,000 CFA (180 US Dollars) in 47.7%. The patient delay was between 31 - 60 days in 61.3% with an average of 58.1 ± 19.4 days (median = 58.2 days); the medical (health) delay was between 31 - 60 days in 62.8% with a mean of 57.2 ± 19.9 days (median = 57.1 days); the diagnostic (total) delay was between 120 - 150 days (4 - 5 months) in 69.6% with a mean of 115.0 ± 23.5 days (median = 114.2 days) ~4 months (Table 2). Comparing means diagnosis delay and cost of expenses before diagnosis. There was a statistically significant difference between the mean of the diagnosis delay and the cost. Those who spent less than 50,000 CFA had a mean delay of 44.4 ± 19.9 days; patients who spent 50,000 - 100,000 CFA had a mean delay of 56.53 ± 16.72 and those who spent more than 100,000 CFA had a mean delay of 61.5 ± 20.6 (p < 0. 00001). Among the 112 smokers, 50.0% (65/112) had spent more than 100,000 FCFA before TB diagnosis; a statistically significant association was observed between cost and smoking, p = 0.009 (Table 3). Comparing mean diagnosis delay between patients who seek or not for traditional medicine consultation, there was a statistically significant difference between the mean patient delays, there were more days for patients who visited traditional therapy 6.84 ± 2.35 95% CI (2.19 - 11.48), p = 0.004. This difference affected the diagnosis delay with a difference of 10.04 ± 2.83 95% CI (4.43 - 15.65) days, p < 0.000001. Univariate analysis was performed (Table 4), primary healthcare/pharmacy [OR = 0.35 (0.20 - 0.60), p < 0.0001], traditional therapy or Self-medication [OR = 2.24 (1.30 - 3.83), p = 0.004], Hemoptysis [OR = 16.07 (2.16 - 119.78), p < 0.0001], Dyspnea [OR = 6.66 (3.72 - 11.92), p < 0.00001], Financial reason [OR = 2.17 (1.18 - 4.01), p = 0.015], Social reason [2.83 (1.46 - 5.51), p = 0.002] and Health

There is a statistically significant association between smoking and non-smoking patients on the cost spent before the diagnosis of tuberculosis (Chi 2 test = 9.54; p = 0.009).

Education level below university, social reasons, and non-request of TB test by the health personnel were identified as independent risk factors for a delayed diagnosis of more than 100 days.

worker did not request for TB test [2.17 (1.17 - 4.04), p = 0.018] were factors associated with a diagnosis delay longer than 100 days. Logistic regression was performed to determine risk factors. Education level below university [aOR = 9.7 (1.9 - 50.2), p = 0.007]; social reasons [aOR = 3.4 (1.2 - 9.4), p = 0.021] and the non-request of TB test by health workers [aOR = 8.1 (2.8 - 22.9), p < 0.0001] were independent risk factors associated with long delay of more than 100 days before diagnosis of tuberculosis (Table 5).

This study has shown that the delay in diagnosis of tuberculosis is relatively very long, around four months in Mali. Several modifiable factors are associated with this delay, education below university, social reasons, and most importantly, the non-request of the diagnostic test by health workers. It can be drawn that patients are visiting late, and TB suggestive signs are not considered at the first consultation. To achieve the objective of 90% screening in Mali, a nationwide survey must be conducted by health workers to identify the reasons for medical and patient delays for TB diagnosis.

The study team sincerely thanks all the patients, physicians, and nurses at the TB treatment Centers in Bamako for their participation.

DS, KO, BB, TK, CMS, SK, IG, LC, YK, APD, FS, SK, DK, OY, GB, and YT have designed the study. DS, KO, BB, TK, CMS, SK, YK wrote the first draft, FS, SK, DK, OY, GB, and YT made substantial corrections, and all the co-authors have read and approved the final version of the manuscript.

This work was supported by the Northwestern University’s Institute for Global Health Catalyzer program, the National Institutes of Health (D43TW010350 and R21AI148033). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

The authors declare no conflicts of interest regarding the publication of this paper.

Soumare, D., Baya, B., Ouattara, K., Kanoute, T., Sy, C.M., Karembé, S., Guindo, I., Coulibaly, L., Kamian, Y., Dakouo, A.P., Sidibe, F., Koné, S., Kone, D., Yossi, O., Berthe, G. and Toloba, Y. (2022) Identifying Risk Factors for Pulmonary Tuberculosis Diagnosis Delays in Mali a West-African Endemic Country. Journal of Tuberculosis Research, 10, 45-59. https://doi.org/10.4236/jtr.2022.101004