texto em 
Inglês (pdf)
Artigo em XML
Referências do artigo
Tradução automática
Permalink
INTRODUCTION
According to the World Health Organization (WHO), tuberculosis (TB) is the most curable infectious disease in the world1. In 2020, TB affected about 9.9 million people worldwide1. Brazil is among the group of 22 countries responsible for 90% of TB cases1-3. In the country, an average of 68,000 new TB cases and around 4,000 deaths are recorded2,3 per year. The proportions of abandonment may contribute to this epidemiological scenario that is difficult to control4,5. In 2020, an abandonment rate of 12.9% was recorded in Brazil, 2.6 times higher than the 5% target set by the WHO1,3.
Identifying the profile of TB patients vulnerable to abandonment of tuberculosis treatment and the factors that can trigger such an outcome becomes an essential tool for establishing actions and strategies involving the management and organization of services, aiming at adherence and control of the disease since abandonment helps in the cycle of spread and contagion of the disease, increased costs, drug resistance, morbidity, and mortality1-6. Given this problem, this study aims to analyze the factors associated with the abandonment of TB treatment.
METHODS
Study design and context
This is a cross-sectional study of individuals diagnosed with TB in Brazil between 2014 and 2019 whose cases had been reported to the Notifiable Diseases Information System (SINAN) of Brazil, in which TB notification is mandatory and its feeding is regularly carried out by all federated entities with information from individuals with suspected or diagnosed diseases and conditions of interest7.
Tuberculosis treatment sensitive to standard drugs lasts at least six months, with a minimum follow-up period which may be extended due to associated diseases or clinical evolution. Thus, individuals are followed longitudinally throughout the treatment, in which periodic information is obtained at different times2,7.
Participants
The study included individuals over 18 years of age whose diagnoses of tuberculosis sensitive to the drugs used in the treatment cases had been reported to SINAN, following national recommendations2.
Individuals with postmortem notification and treatment termination status ignored or blank were excluded, as were those with information on changes in regimen, treatment failure, change in diagnosis, drug-resistant tuberculosis, and death from tuberculosis or other causes.
Variables
The situation at the end of treatment was the dependent variable obtained from the tuberculosis follow-up report at the end of treatment, in which it was reclassified as abandonment of treatment (abandonment and primary abandonment) and cure.
b) Contextual, level II: Geographic region (southeast/northeast/central-west/south/north); Area of residence (urban/rural/peri-urban); Population deprived of liberty (no/yes); Healthcare professional (no/yes); Homeless population (no/yes); Immigrants (no/yes); Beneficiary of income transfer or government aid (no/yes).
c) Associated diseases, level III: Human immunodeficiency virus infection/acquired immunodeficiency syndrome (no/yes); Alcoholism (no/yes); Illicit drug use (no/yes); Smoking (no/yes); Diabetes mellitus (no/yes); Mental disorder (no/yes).
d) Current tuberculosis treatment, level IV: Type of notification (new cases/recurrence/resumption after abandonment/transfer/does not know); Clinical form of tuberculosis (pulmonary/extrapulmonary/pulmonary + extrapulmonary); Sputum smear microscopy results (negative/positive/not performed); Initial culture results (negative/positive/progress/not performed); Chest X-ray (not suggestive of tuberculosis/suggestive of tuberculosis); Directly observed treatment (no/yes).
Data Sources
Data on individuals with tuberculosis were obtained from the Notifiable Diseases Information System (SINAN) provided by the National Tuberculosis Control Program.
Statistical methods
Initially, the absolute and relative frequencies of TB treatment were calculated for each explanatory variable. In the bivariate analysis, Poisson regression of robust variance was used to obtain measures of association between the explanatory variables and the abandonment of TB treatment.
The variables with a p-value <0.20 in the bivariate analysis were introduced into the Poisson regression model of robust variance according to the hierarchical levels proposed by Maciel and Reis-Santos: level I (sociodemographic), level II (contextual), level III (associated diseases/comorbidities) and level IV (current clinical situation related to TB)8. The variables were maintained at the following levels as an adjustment in the model if the p-value was < 0.05. The association of each factor of abandonment of TB treatment is interpreted as adjusted for the variables of the hierarchical levels above it and also of the same level.
The results were expressed by the measure of the association of prevalence ratio (PR) and 95% confidence intervals (95%CI). Statistical analyses were performed using Stata v. 14.0 (StataCorp, CollegeStation, TX, USA).
RESULTS
Between 2014 and 2019, 508,787 cases of tuberculosis were reported in individuals over 18 years of age in Brazil. A total of 18,142 (3.5%) cases of death from tuberculosis, 21,366 (4.2%) cases of death from other causes, 31,940 (6.2%) cases of transfers, 3,542 (0.7%) due to change of diagnosis, 5,199 (1.1%) due to drug-resistant tuberculosis, 2,690 (0.5%) due to change of treatment regimen, and 373 (0.1%) due to treatment failure, were excluded from the study. The final population consisted of 364,440 individuals, of whom 59,871 (16.4%) abandoned tuberculosis treatment.
Tables 1 and 2 show the distribution of the frequency of abandonment of tuberculosis treatment by the variables analyzed in the study. In the hierarchical analysis, shown in tables 3 and 4, the prevalence ratio of the abandonment was higher in homeless individuals (PR 2.75; 95% CI 2.10-3.61), resumed treatment after the abandonment (PR 1.91; 95% CI 1.54-2.37), illicit drug use (PR 1.85; 95% CI 1.49-2.28), black race/skin color when compared to white race/color (PR 1.79; 95% CI 1.46-2.20), individuals with HIV/AIDS (PR 1.59; 95% CI 1.30-1.93), and alcoholics (PR 1.38; 95% CI 1.14-1.68). On the other hand, individuals with more than 8 years of schooling (PR 0.53; 95% CI 0.33-0.84), extrapulmonary TB (PR 0.54; 95% CI 0.38-0.77), and those who underwent DOT (PR 0.78; 95%CI 0.66-0.91) had the lowest prevalence ratios.
Table 1 : Distribution of the frequency of abandonment of tuberculosis treatment by demographic and contextual characteristics, Brazil, 2014 to 2019
| Abandonment | ||
|---|---|---|
| Yes | No | |
| Variable | N (%) | N (%) |
| Sex (n = 364,407) | ||
| Female | 14,338 (13.09) | 95,542 (86.91) |
| Male | 45,475 (17.87) | 209,002 (82.13) |
| Age (n = 364,374) | ||
| 18-19 years | 2,228 (17.24) | 10,696 (82.76) |
| 20-39 | 36,237 (19.67) | 147,987 (80.33) |
| 40-59 | 17,644 (14.84) | 101,278 (85.16) |
| ≥ 60 | 3,756 (7.78) | 44,548 (92.22) |
| Race/skin color (n = 337,599) | ||
| White | 14,989 (13.12) | 99,247 (86.88) |
| Black | 10,260 (21.72) | 36,970 (78.28) |
| Brown (Parda) | 29,268 (17.20) | 140,887 (82.80) |
| Asians/Indigenous | 728 (12.18) | 5,250 (87.82) |
| Years of study (n = 273,390) | ||
| Illiterate | 2,401 (15.19) | 13,404 (84.81) |
| 0-4 | 10,306 (17.15) | 49,794 (82.85) |
| 5-8 | 25,157 (17.94) | 115,043 (82.06) |
| > 8 | 5,116 (8.93) | 52,169 (91.07) |
| Geographic region (n = 364,440) | ||
| Southeast | 28,820 (16.36) | 147,375 (83.64) |
| South | 8,084 (18.33) | 36,013 (81.67) |
| Central-West | 2,708 (16.75) | 13,456 (83.25) |
| Northeast | 14,054 (15.68) | 75,569 (84.32) |
| North | 6,205 (16.18) | 32,156 (83.82) |
| Area of residence (n = 256,954) | ||
| Urban | 40,635 (17.73) | 188,588 (82.27) |
| Rural | 2,645 (10.58) | 22,359 (89.42) |
| Peri-urban | 421 (15.44) | 2,306 (84.56) |
| Income transfer benefit (n = 180,092) | ||
| Yes | 2,172 (14.24) | 13,080 (85.76) |
| No | 28,103 (17.05) | 136,737 (82.95) |
| Population deprived of liberty (n = 313,833) | ||
| Yes | 4,237 (11.20) | 33,606 (88.80) |
| No | 47,062 (17.05) | 228,928 (82.95) |
| Healthcare professional (n = 302,062) | ||
| Yes | 254 (5.71) | 4,196 (94.29) |
| No | 49,410 (16.60) | 248,202 (83.40) |
| Homeless population (n = 312,120) | ||
| Yes | 6,261 (54.39) | 5,250 (45.61) |
| No | 44,919 (14.94) | 255,690 (85.06) |
| Immigrants (n = 259,013) | ||
| Yes | 325 (21.10) | 1,215 (78.90) |
| No | 43,254 (16.80) | 214,219 (83.20) |
Table 2 : Distribution of the frequency of abandonment of tuberculosis treatment by comorbidities and clinical status of tuberculosis treatment. Brazil, 2014 to 2019
| Abandonment | ||
|---|---|---|
| Yes | No | |
| Variable | N (%) | N (%) |
| HIV / AIDS (n = 326,347) | ||
| Yes | 9,153 (30.70) | 20,660 (69.30) |
| No | 43,311 (14.61) | 253,223 (85.39) |
| Alcoholism (n = 343,115) | ||
| Yes | 18,196 (27.72) | 47,435 (72.28) |
| No | 37,484 (13.51) | 240,000 (86.49) |
| Smoking (n = 311,290) | ||
| Yes | 17,120 (23.07) | 57,101 (76.93) |
| No | 33,390 (14.04) | 204,359 (85.96) |
| Illicit drug use (n = 310,839) | ||
| Yes | 17,633 (35.53) | 31,994 (64.47) |
| No | 32,841 (12.57) | 228,371 (87.43) |
| Diabetes mellitus (n = 341,671) | ||
| Yes | 2,557 (9.54) | 24,223 (90.45) |
| No | 52,540 (16.69) | 262,351 (83.31) |
| Mental disorder (n = 341,016) | ||
| Yes | 1,707 (20.61) | 6,575 (79.39) |
| No | 53,242 (16.00) | 279,492 (84.00) |
| Type of notification (n = 364,440) | ||
| New case | 38,833 (13.00) | 259,919 (87.00) |
| Recurrence | 4,493 (16.68) | 22,449 (83.32) |
| Resumption after abandonment | 15,115 (49.99) | 15,123 (50.01) |
| Does not know | 155 (22.33) | 539 (77.67) |
| Transfer | 1,273 (16.30) | 6,536 (83.70) |
| Clinical form (n = 364,432) | ||
| Pulmonary | 53,643 (17.14) | 259,363 (82.86) |
| Extrapulmonary | 4,633 (10.91) | 37,831 (89.09) |
| Pulmonary + Extrapulmonary | 1,594 (17.79) | 7,368 (82.21) |
| Sputum smear results (n = 355,090) | ||
| Negative | 10,876 (14.94) | 61,937 (85.06) |
| Positive | 32,591 (16.59) | 163,906 (83.41) |
| Not Performed | 14,977 (17.46) | 70,803 (82.54) |
| Result of the initial culture (n = 364,440) | ||
| Negative | 4,609 (12.36) | 32,677(87.64) |
| Positive | 14,312 (17.30) | 68,429 (82.70) |
| In Progress | 1,304 (20.09) | 5,186 (79.91) |
| Not Performed | 39,646 (16.66) | 198,277 (83.34) |
| Chest X-ray (n = 348,306) | ||
| Not suggestive of TB | 2,621 (12.31) | 18,678 (87.69) |
| Suggestive of TB | 41,163 (16.57) | 207,184 (83.43) |
| Not Performed | 13,486 (17.14) | 65,174 (82.86) |
| DOT (n = 129,037) | ||
| Yes | 12,725 (13.23) | 83,432 (86.77) |
| No | 6,135 (18.66) | 26,745 (81.34) |
HIV/AIDS: human immunodeficiency virus infection/acquired immunodeficiency syndrome; TB: tuberculosis; DOT: directly observed treatment.
Table 3 : Raw and adjusted prevalence ratio of individuals who abandoned tuberculosis treatment by sociodemographic and contextual characteristics, Brazil, 2014 to 2019
| Raw Analysis | Hierarchical Analysis | |||
|---|---|---|---|---|
| Variables | PR (95%CI) | P value* | PR (95%CI) | P value* |
| Sociodemographic – Level I | ||||
| Sex | < 0.001 | 0.078 | ||
| Female | 1.00 | 1.00 | ||
| Male | 1.36 (1.34-1.38) | 1.17 (0.98-1.39) | ||
| Age | < 0.001 | 0.001 | ||
| 18-19 years | 1.00 | 1.00 | ||
| 20-39 | 1.14 (1.09-1.18) | 1.04 (0.70-1.56) | ||
| 40-59 | 0.86 (0.82-0.89) | 0.75 (0.49-1.14) | ||
| ≥ 60 | 0.45 (0.42-0.47) | 0.38 (0.23-0.65) | ||
| Race/skin color | < 0.001 | < 0.001 | ||
| White | 1.00 | 1.00 | ||
| Black | 1.65 (1.61-1.69) | 1.79 (1.46-2.20) | ||
| Brown (Parda) | 1.31 (1.28-1.33) | 1.12 (0.93-1.34) | ||
| Asians/Indigenous | 0.92 (0.86-0.99) | 0.11 (0.01-0.84) | ||
| Years of study | < 0.001 | < 0.001 | ||
| Illiterate | 1.00 | 1.00 | ||
| 0-4 | 1.12 (1.08-1.17) | 1.21 (0.80-1.83) | ||
| 5-8 | 1.18 (1.13-1.22) | 1.21 (0.81-1.82) | ||
| > 8 | 0.58 (0.56-0.61) | 0.53 (0.33-0.84) | ||
| Contextual – Level II | ||||
| Geographic Region | < 0.001 | < 0.001 | ||
| Southeast | 1.00 | 1.00 | ||
| South | 1.12 (1.09-1.14) | 0.67 (0.54-0.81) | ||
| Central-West | 1.02 (0.98-1.06) | 0.33 (0.21-0.54) | ||
| Northeast | 0.95 (0.94-0.97) | 0.69 (0.54-0.89) | ||
| North | 0.98 (0.96-1.01) | 0.39 (0.27-0.58) | ||
| Area of residence | < 0.001 | < 0.001 | ||
| Urban | 1.00 | 1.00 | ||
| Rural | 0.59 (0.57-0.61) | 0.42 (0.26-0.67) | ||
| Peri-urban | 0.87 (0.79-0.95) | 1.03 (0.42-2.53) | ||
| Income transfer benefit | < 0.001 | 0.112 | ||
| Yes | 0.83 (0.80-0.86) | 0.61 (0.33-1.11) | ||
| No | 1.00 | 1.00 | ||
| Population deprived of liberty | < 0.001 | 0.032 | ||
| Yes | 0.65 (0.63-0.67) | 1.35 (1.02-1.78) | ||
| No | 1.00 | 1.00 | ||
| Healthcare Professional | < 0.001 | 0.800 | ||
| Yes | 0.34 (0.30-0.38) | 0.84 (0.21-3.21) | ||
| No | 1.00 | 1.00 | ||
| Homeless Population | < 0.001 | < 0.001 | ||
| Yes | 3.64 (3.57-3.70) | 2.75 (2.10-3.61) | ||
| No | 1.00 | 1.00 | ||
| Immigrants | < 0.001 | 0.450 | ||
| Yes | 1.25 (1.14-1.38) | 1.85 (0.37-9.18) | ||
| No | 1.00 | 1.00 | ||
| *Poisson regression; PR: prevalence ratio; 95%CI: 95% confidence interval. | ||||
Table 4 : Raw and adjusted prevalence ratio of individuals who abandoned tuberculosis treatment by clinical characteristics and comorbidities, Brazil, 2014 to 2019.
| Raw Analysis | Hierarchical Analysis | |||
|---|---|---|---|---|
| Variables | PR (95%CI) | P value* | PR (95%CI) | P value* |
| Associated diseases – Level III | ||||
| HIV/AIDS | < 0.001 | < 0.001 | ||
| Yes | 2.10 (2.06-2.14) | 1.59 (1.30-1.93) | ||
| No | 1.00 | 1.00 | ||
| Alcoholism | < 0.001 | 0.001 | ||
| Yes | 2.05 (2.02-2.08) | 1.38 (1.14-1.68) | ||
| No | 1.00 | 1.00 | ||
| Smoking | < 0.001 | 0.012 | ||
| Yes | 1.64 (1.61-1.66) | 0.72 (0.56-0.93) | ||
| No | 1.00 | 1.00 | ||
| Illicit drug use | < 0.001 | < 0.001 | ||
| Yes | 2.82 (2.78-2.87) | 1.85 (1.49-2.28) | ||
| No | 1.00 | 1.00 | ||
| Diabetes mellitus | < 0.001 | 0.366 | ||
| Yes | 0.57 (0.55-0.59) | 0.82 (0.53-1.25) | ||
| No | 1.00 | 1.0 | ||
| Mental disorder | < 0.001 | 0.433 | ||
| Yes | 1.28 (1.23-1.34) | 0.82 (0.51-1.33) | ||
| No | 1.00 | 1.0 | ||
| Current tuberculosis treatment - Level IV | ||||
| Type of Notification | < 0.001 | < 0.001 | ||
| New Case | 1.00 | 1.00 | ||
| Recurrence | 1.28 (1.24-1.31) | 0.90 (0.65-1.26) | ||
| Resumption after abandonment | 3.84 (3.78-3.90) | 1.91 (1.54-2.37) | ||
| Does not know | 1.71 (1.49-1.97) | 2.57 (0.70-9.43) | ||
| Transfer | 1.25 (1.19-1.31) | 0.82 (0.40-1.70) | ||
| Clinical Form | < 0.001 | < 0.001 | ||
| Pulmonary | 1.00 | 1.00 | ||
| Extrapulmonary | 0.63(0.61-0.65) | 0.54 (0.38-0.77) | ||
| Pulmonary + Extrapulmonary | 1.03 (0.99-1.08) | 0.85 (0.57-1.26) | ||
| Sputum smear results | < 0.001 | 0.240 | ||
| Negative | 1.00 | 1.00 | ||
| Positive | 1.11 (1.08-1.13) | 1.09 (0.88-1.33) | ||
| Not Performed | 1.16 (1.14-1.19) | 1.25 (0.96-1.63) | ||
| Results of the initial culture | < 0.001 | 0.085 | ||
| Negative | 1.00 | 1.00 | ||
| Positive | 1.39 (1.35-1.44) | 1.40 (1.01-1.95) | ||
| In Progress | 1.62 (1.53-1.71) | 1.57 (0.93-2.64) | ||
| Not Performed | 1.34 (1.31-1.38) | 1.49 (1.09-2.05) | ||
| Chest X-ray | < 0.001 | 0.623 | ||
| Not suggestive of TB | 1.00 | 1.00 | ||
| Suggestive of TB | 1.34 (1.29-1.39) | 0.87 (0.58-1.31) | ||
| Not Performed | 1.39 (1.33-1.44) | 0.96 (0.61-1.49) | ||
| DOT | < 0.001 | 0.002 | ||
| Yes | 0.70 (0.68-0.72) | 0.78 (0.66-0.91) | ||
| No | 1.00 | 1.00 | ||
*Poisson regression; PR: prevalence ratio; 95% CI: 95% confidence interval; HIV/AIDS: human immunodeficiency virus infection/acquired immunodeficiency syndrome; TB: tuberculosis; DOT: directly observed treatment.
DISCUSSION
The prevalence of the abandonment of tuberculosis treatment in Brazil between 2014 and 2019 was 16.4%, higher than the WHO recommendation of 5%. Self-declared individuals of black race/color, homeless, people living with HIV, alcoholics, illicit drug users, and resumption after the abandonment had the highest prevalence of ATT. On the other hand, individuals who received government assistance, extrapulmonary TB, and who underwent directly observed treatment had the lowest.
The information is generated by several health services, and, although the Ministry of Health standardizes the completion of the information system, the possibility of a classification different from the one recommended cannot be ruled out. However, it is believed that these limitations did not interfere with the results presented, an understanding reinforced by prior evaluations that demonstrated the quality of SINAN, and by the consistency of these results with the findings of the accumulated literature.
It was observed that black individuals had the highest prevalence of abandonment. This occurred because abandonment is linked to the social and economic vulnerability of the black population in Brazil which has less access to employment opportunities, lower income, precarious housing, and difficulty in accessing education and food6,8-10.
Another factor linked to social vulnerability is the abuse of alcohol and other drugs. Precarious living conditions and increased risk of hepatotoxicity due to tuberculosis treatment in this group of patients may aggravate the disease and impair the therapeutic regimen, increasing the chance of abandonment11-13. In addition, alcohol and other drug abuse is a survival characteristic in homeless individuals13,14. Dependence on licit or illicit substances, the presence of other comorbidities such as HIV infection, social marginalization, and low access to public services hinder the process of caring for this population. The results show that the homeless population is especially vulnerable to the abandonment of tuberculosis treatment14-16.
People living with HIV also live in social and economic vulnerability, which influences adherence to TB treatment, as well as the use of antiretrovirals. Abandonment of antiretroviral use was associated with social and economic issues in a systematic review that assessed adherence to treatment in people living with HIV17. Given this, the need to intensify the follow-up of cases of TB-HIV co-infection is reinforced, as well as the incorporation of HIV serological testing in individuals diagnosed with TB18.
The prevalence of abandonment was higher among individuals who resumed treatment. This result shows that even though treatment and access to diagnostic tests are offered, the failure to reduce social and economic vulnerabilities in the daily lives of these individuals makes them more likely to abandon treatment again.
Directly observed treatment had the lowest prevalence of abandonment of TB treatment. Monitoring daily medication intake is a strategy to protect against abandonment. The increased coverage of Family Health Strategy services strengthens the practice of DOT, with a positive impact on longitudinal care centered on individuals with TB19,20.
The lowest prevalence of abandonment of TB treatment was among individuals with extrapulmonary disease. Corroborating our results, studies conducted with SINAN data describe the extrapulmonary form as a factor less likely to abandon treatment21-23. However, we suggest caution in the interpretation of these results, as the extrapulmonary form is more prone to other unfavorable outcomes, such as death and treatment failure24.
As shown, social and economic vulnerability is present among individuals affected by tuberculosis and is associated with the abandonment of tuberculosis treatment. Therefore, it is necessary to implement educational strategies in health, implement street clinics with a greater focus on tuberculosis treatment, and increase the supply of directly observed treatment to reduce abandonment.
