Temporal pattern of tuberculosis cure, mortality, and treatment abandonment in Brazilian capitals

Objective: to analyze the temporal pattern of tuberculosis cure, mortality, treatment abandonment in Brazilian capitals. Method: this is an ecological study whose data source was the Information System of Notifiable Diseases for Tuberculosis (Sistema de Informação de Agravos de Notificação para Tuberculose). For analysis of temporal evolution, regressions by join points were performed considering the annual percentage variation and the significance of the trend change with 95% confidence interval. Results: 542,656 cases of tuberculosis were found, with emphasis on a 3% decrease per year in the cure rate for Campo Grande (interval: −5.0 - −0.9) and a 3.5% increase for Rio de Janeiro (interval: 1.9 - 4.7). Regarding abandonment, it decreased 10.9% per year in Rio Branco (interval: −15.8 - −5.7) and increased 12.8% per year in Fortaleza (interval: 7.6 - 18.3). For mortality, a decreasing or stationary tendency was identified, with a greater decrease (7.8%) for Porto Velho (interval:−11.0 - −5.0) and a lower one (2.5%) in Porto Alegre (interval:−4.5 - −0.6). Conclusion: the rates of cure and abandonment are far from the ones recommended by the World Health Organization, showing that Brazilian capitals need interventions aimed at changing this pattern.


Introduction
Worldwide, tuberculosis (TB) is the ninth cause of death and the first one by infectious disease. In 2017,

million new cases and 1.3 million deaths caused by this infection were estimated, showing a reality that is inconsistent with what is foreseen in the Sustainable
Development Goals (SDG), which guide the reduction of the TB epidemic (1)(2) .
Specifically for TB, the SDG recommend prioritizing the incidence reduction by 80% and its deaths by 90% up to 2030. In addition to the SDGs aimed at reducing the outcomes related to TB, we have the End TB Strategy, created by the World Health Organization (WHO), which expands this deadline to 2035, expanding the incidence reduction to 90% and deaths to 95% (2)(3)(4)(5) . The WHO also recommends a minimum 85% cure and a maximum 5% treatment abandonment (6)(7) .
The End TB Strategy is based on the principles of "integrated care and prevention, focused on the patient; bold policies and support systems, with emphasis on the social protection of vulnerable populations; and intensification of research and innovation." However, obstacles that impede the achievement of the proposed goals regarding TB still have to be overcome, especially in emerging countries. In this perspective, in 2017, Brazil signed a commitment term to develop a research agenda in the area of TB (4)(5) .
Brazil is on the list of countries that amount to 49% of the world TB burden and 60% of its resistant forms, and, even with an annual decrease of 1.7% in the TB incidence coefficient, it currently has a national average of 32.4/100,000 inhabitants (8) , which makes the constant verification of TB data urgent. Thus, this study presents a large temporal series was performed for the outcomes (cure, abandonment, and mortality) by TB in Brazilian capitals, by their population aggregation and data completeness in the bases, which makes the evidence generated have greater external validity. In addition, clusters are determinant in the TB transmission chain (9) .
One can find in the literature studies on TB temporal patterns. However, they address only one capital (or specific region) and mainly investigate incidence or prevalence (10)(11)(12)(13) . This investigation includes three important outcomes (cure, abandonment, and mortality), which are seen in a large temporality, population, and statistical sensitivity.
Studies with these characteristics have been pointed out as important (10)(11)(12)(13) , as they may show results that fill gaps, such as those related to Brazil achieving the goals set to reduce negative outcomes by TB. In this study, one can have a broad view on how cure and abandonment may affect mortality.
The literature endorses the need to fill knowledge gaps about tuberculosis treatment outcomes as a strategy for nursing care and health management, mainly focused on people with a chance to present negative results (10) . Thus, this study aimed to analyze the temporal pattern of the outcomes cure, abandonment, and mortality by TB in Brazilian capitals.

Method
This is an ecological study of time series, which analyzed three outcomes of TB treatment ( The data for the systems previously exposed were on. The data were made available online and the files were downloaded in CSV format. The raw data of the disease in each year and its outcomes were tabulated in Excel spreadsheet and imported into the free software Joinpoint Regression Program version 4.6.0.0 (14) . This software was created to analyze the tendency of cancer, but is currently used in other fields of epidemiology, due to its statistical potentiality in analyzing temporal patterns (11,(15)(16) .
The analysis that the program performs is an increasing trend (11,(15)(16) . At the end of the period, it was possible to obtain the Average Annual Percentage Change (AAPC), which shows how the change occurred during the period studied. If there is more than one inflection point, the AAPC will consider all of them for its calculation, otherwise the AAPC will be equal to the APC. adjusted assuming that the number of inflection points could vary from zero (only one segment) to two (three segments) over the years. A 5% significance level was established to test the null hypothesis that the APC and the AAPC of the series are equal to zero (17) . Thus, both for the analysis of APC and AAPC, the results with p<0.05 or only positive (increasing trend) or only negative (decreasing trend) 95%CI are significant.
The year of occurrence was defined as independent variable and the percentage of people with the disease in each year was defined as dependent variable, both calculated directly in the program and standardized according to the previously exposed criteria of logarithms.
For this, the number of cases was included with the selected outcome as numerator and the total number of cases as denominator. However, for the outcome mortality, the number of deaths by TB was selected as numerator and the population of the year was used as denominator.
The values of cure and abandonment were analyzed in percentage and, for mortality values, the coefficient was considered for 100,000 inhabitants. The presence of lacking data was an exclusion factor in these analyses. in each year can be seen in Table 1.
For the analysis of the trends of the three outcome situations analyzed,the AAPC at the end of the periods was considered. After this analysis, it was possible to identify an expressive heterogeneity throughout the national territory, even in states belonging to the same region. Nationwide, 22% (6) capitals showed a significant increase in the annual cure rate, and 15%  Table 2).
The next outcome, treatment abandonment, presented expressive increasing rates. Of the 26 cities studied (Palmas could not be analyzed due to lack of data in some years during the series), half (13) significantly increased the abandonment rate and only 4% (1) decreased this rate over the years. Of these capitals, 46% (12)

Discussion
The main results lead us to the finding that the cure is stationary, while treatment abandonment showed a growth trend and mortality, a decrease trend.
The limitation of this study occurs mainly because of its secondary-based situation, which may restrict the external validity of the data, and by the change in the notification and follow-up form of the disease in 2014.
Concerning the interpretation of results related to cure, the national database study (18) suggests that TB cure rates have decreased, even in the presence of new interventions, such as the inclusion of ethambutol (E) in the former rifampicin, isoniazid, pyrazinamide (RHZ) treatment. New cases present better cure rates compared to retreatment or return after abandonment (19) . The literature points out that knowledge, ease of access to the health service, adequate infrastructure, health promote the cure (20)(21) .
Regarding the interpretation of results related to treatment abandonment, this study indicates that the reduction of cure is intimately associated with increased treatment abandonment (18) . The following factors are associated with treatment abandonment: male sex; retreatment; use of illegal drugs; malnutrition; few years of schooling; lack of knowledge about the disease; access to health service (22)(23)(24) ; and coinfection by the Human Immunodeficiency Virus (HIV). This decreases by 58% the chances of cure of a TB patient, increases by 50% the chances of abandonment and by 94% the chances of death by TB (12) , increasing the notifications in the information systems (25) . It was shown that, as much as the abandonment rate is within the one recommended by the WHO, the cases of coinfection must be monitored to avoid new negative outcomes (26) .
Moreover, the expansion of drug treatment coverage to people living with HIV, along with preventive measures, has the potential to reduce the TB burden (27) .
This study has shown a significant increase in the abandonment of TB treatment in most national capitals.
As pointed out in the literature, a factor that may contribute to this outcome is the lack of qualified and structured care, because the search for more distant units, but that are references in TB care, is still common among patients, who travel longer itineraries to develop their treatment (28) . It is believed that the decentralization of health services specialized in caring for people with TB promotes quality care and the supervised treatment of patients, which could decrease abandonment.
In addition, cases classified as retreatment have twice as much chance of new abandonment, as well as those with negative bacilloscopy and unknown HIV status (29) .
The literature indicates that the reduction of extreme poverty would decrease the incidence of TB by 33.4% up to 2035, and that the increase in social protection would reduce it by 76.1% up to the same year (30) .
In this context, it is worth noting the importance of the Directly Observed Therapy (DOT) in the TB care.
The DOT, component of the End TB Strategy, is another approach in the treatment of people with TB, especially those with more chances of abandoning treatment, as it accompanies the taking of medications (daily or alternating), a resource used to expand the adherence to drug treatment (31) . According to Datasus, 20 of the 27 capitals studied had a coverage of DOT lower than 50%, data that may be a possible explanation for the growth in the numbers of abandonment in Brazil. Even so, although the DOT approximates patient and health sector, there is still difficulty in shaping its real impact on the success of TB treatment (32) . there is a need for strategies to reduce treatment abandonment and drug resistance (33) .
Regarding mortality, it had decreased across the world in people without HIV infection (34) , and has the following associated factors: male sex; lower economic power; lower Human Development Index (HDI); higher immigration rate; black ethnicity and higher coinfection coefficients for HIV (10,13,(35)(36) ; vulnerable (immunosuppressed) populations, people with circulatory problems and neoplasia; and people with delayed diagnosis and inefficient follow-up (37) . The decrease in mortality observed in this study may derive from public policies that prioritize the monitoring of the patient, free treatment in the appropriate time, and social protection measures with money transfer (38) . The investigation of all national capitals has shown data for analysis, which can subsidize new policies to increase cure and reduce unfavorable outcomes.

Conclusion
When analyzing the temporal pattern of the outcomes cure, abandonment, and mortality by TB in