Epidemiological aspects of HIV infection and AIDS among indigenous populations

ABSTRACT OBJECTIVE To describe the epidemiological aspects of HIV infection and AIDS among indigenous peoples of the state of Mato Grosso do Sul, Brazil. METHODS This is a descriptive epidemiological study on the occurrence and distribution of HIV infection and AIDS in the indigenous population assisted by the Distrito Sanitário Especial Indígena (Indigenous Special Health District) Mato Grosso do Sul between 2001 and 2014, based on three secondary databases. Annual rates of HIV and AIDS detection and prevalence were calculated, considering case distribution according to village, Health Base Pole and sociodemographic variables. Accumulated rates of detection, mortality and case fatality were calculated by ethnic group and for the Health Base Pole with the highest number of cases. RESULTS The HIV detection rate fluctuated between 0.0 and 18.0/100 thousand people in the study period. For AIDS, there was no notification before 2007, but in 2012 its rate reached 16.6/100 thousand. HIV prevalence grew between 2001 and 2011, and it continuously grew for AIDS starting from 2007. The highest HIV detection rates occurred among Guarani peoples (167.1/100 thousand) and for AIDS, among the Kaiowá peoples (79.3/100 thousand); mortality and fatality rates were higher among the Kaiowá. Regarding the Dourados Health Base Pole, the AIDS detection rate increased, and the mortality and fatality rates decreased. CONCLUSIONS HIV infection and AIDS have been increasing among indigenous peoples, with distribution of the disease mainly in the Health Base Poles of the southern region of the state, where greater economic and social vulnerability are also observed. The endemic character of HIV and AIDS can become epidemic in some years given the existence of cases in other villages in the state. Its occurrence among the Guarani and Kaiowá populations indicates the need for expanded diagnosis, access to treatment and prevention measures.


INTRODUCTION
Acquired immunodeficiency syndrome (AIDS) has already affected millions of people worldwide. By 2015, about 36.7 million people were living with the virus a . In Brazil, from 1980 to June 2017, 882,810 cases of AIDS were recorded b . Between 2006 and 2016, there was a trend towards the stabilization of the AIDS detection rate, considering a 18.5/100 thousand inhabitants mean in 2016. However, in the state of Mato Grosso do Sul, an 8.2% growth occurred in the period, represented by a 19.8/100 thousand inhabitants rate in 2016, the highest in the Midwest region b .
The impact of AIDS among indigenous populations is still little known. However, studies have highlighted the greater vulnerability of these population groups to HIV transmission, related to bad living conditions, lower socioeconomic and educational level, social exclusion and difficulty in accessing health services 1,2 . Information from countries such as the United States, Canada, New Zealand and Australia indicates a significant increase in the HIV detection rate among indigenous peoples in recent decades [3][4][5] . In general, this rate is higher among indigenous individuals than non-indigenous, indicating health inequities in the access to diagnosis and treatment of the disease 1,2,5 .
The first case of AIDS in indigenous populations in Brazil was recorded in 1987, and only 33 cases were known until 1999. This number increased to 1,042 between 2000 and June 2017 b . However, the impact of AIDS on indigenous peoples cannot be characterized by absolute numbers and/or rates only. The sociocultural, economic, demographic, and geographic dimensions of each indigenous people, associated with the process of interaction with non-indigenous society, expose them to unique risk factors that lead to greater vulnerability when compared with other Brazilian population groups 1,6,c .
The provision of health services for indigenous peoples also has repercussions in this scenario. In recent Brazilian history, the health care of indigenous peoples gained strength in 1999 with the creation of the Subsistema de Atenção à Saúde Indígena, within the scope of the Unified Health System (Indigenous Health Care Subsystem -Sasi-SUS) d .
Considering the need for investigations that fill the gaps related to the occurrence of HIV and AIDS among indigenous peoples, and that can subsidize the formulation of appropriate care policies directed to this population, this study aimed to describe the epidemiological aspects of these cases among the indigenous peoples of the state of Mato Grosso do Sul, from 2001 to 2014. The data recorded by the EMSI during the study period were obtained from June 2014 to January 2015 by manually linking the name, date of birth and name of the mother to three databases: 1. records from daily evaluations made by EMSI; 2. the Sinan database; 3. Sistema de Informação de Atenção à Saúde Indígena (Indigenous Health Care Information System -Siasi). The place of residence was considered as the one registered at the time of the HIV diagnosis by the EMSI or the submitted to Sinan for AIDS cases. Sociodemographic variables (ethnic group, sex, date of birth and schooling) were obtained from Siasi, as well as the population bases used for calculating the rates described below.

This
The number of newly recorded cases of each was used to calculate the annual rates of HIV and AIDS detection. People directly diagnosed with AIDS were only considered for the AIDS detection rate. To include people diagnosed in areas unassisted by DSEI-MS, the territory to which the person moved to was considered, as well as the year in which it occurred.
The number of old and new HIV cases of each year was used to calculate HIV prevalence. Deaths, cases that evolved to AIDS, and people who moved from the areas served by DSEI-MS were excluded. The number of old cases and the new diagnoses was used to calculate AIDS prevalence, in addition to those that evolved to AIDS in that year. Deaths and people who moved from the areas served by DSEI-MS were excluded.
The analysis of the association between the years and the detection rate of HIV and AIDS, as well as the prevalence, was performed using Pearson's chi-square test with Bonferroni correction when necessary, using the IBM SPSS  statistical program, version 24.0, considering a 5% significance level.
The other results were presented as descriptive statistics. The spatial distribution of HIV and AIDS cases in the area covered by DSEI-MS was determined according to the place of residence (village and Health Base Pole) at the time of diagnosis. HIV and AIDS detection rates, as well as accumulated AIDS mortality and fatality for the whole period were estimated, considering the total population of each ethnic group in 2014. The accumulated rates were also calculated for the three Health Base Poles that concentrated the most diagnoses, divided in two periods (    Tacuru (99.2/100 thousand) and Iguatemi (97.9/100 thousand), leaving the Dourados Base out of the first three places ( Table 1).
The age group with the highest incidence of HIV and AIDS infection was 20 to 39 years old, with 43 (44.3%) and 22 (44.0%) cases, respectively. There was statistical significance for HIV cases, for a 0.040 p-value. In the distribution by sex, most HIV and AIDS cases occurred among women, corresponding to 60 (61.9%) and 30 (60%) cases, respectively. We note that HIV infection occurred earlier in women than in men, given that in the 10-19 age group the ratio between the sexes was approximately nine women for each man (  Det rt HD: detection rate by Health Base Pole *One (1) case diagnosed in an individual who resided in a municipality at the time of diagnosis and who then moved to the village during the study period was added.
Although the HIV and AIDS detection rates were higher for the Guarani and Terena ethnic groups, we must highlight the Ofaié ethnic group, whose rates were considerably higher than the other when considering their small population size. Mortality and fatality rates were also higher for the Kaiowá when compared to the other ethnic groups. Moreover, deaths   were recorded only for the Guarani and Terena ethnic groups. Although the Terena had the lowest mortality rate, they had the second highest fatality rate ( Table 3).
Comparisons of detection, mortality and fatality rates between the two analyzed periods (2001 to 2007 and 2008 to 2014) showed that the AIDS detection rate increased more than that of HIV infection. Furthermore, the mortality rate by AIDS increased, whereas the fatality decreased. Of the Health Base Poles with highest detection rates of HIV and AIDS cases, Iguatemi was the only one that showed a decrease in both (HIV and AIDS). In contrast, the Amambai Center presented increased detection rates for both HIV and AIDS. The AIDS detection rate increased in the Dourados Center (25.4/100 thousand for 72.4/100 thousand) whereas its mortality and fatality decreased (Table 4).

DISCUSSION
This study showed the evolution of the detection rates of HIV infection and AIDS in the indigenous population of Mato Grosso do Sul. Women in the 20 to 39 years age group, with lower schooling and of the Kaiowá ethnic group presented the highest HIV and AIDS infection percentages. The Kaiowá, Terena and Guarani ethnic groups were the most exposed one, with records of high HIV infection and AIDS rates. These results also contributed to the high mortality and fatality rates in these groups. The Health Base Poles of Amambai, Dourados and Iguatemi, located in the southern region of the state, concentrated around 70% of cases of HIV infection and AIDS.
Oscillations in detection rates in the study period may be more strongly associated with the diagnostic ability of services than to the epidemiological behavior of the virus transmission or disease burden on the population. These must be interpreted as a warning sign for the health care services for the state's indigenous peoples.
The Guarani and Kaiowá have a very complex health picture, with high prevalence of infectious diseases, high infant mortality and malnutrition, as well as the increasing prevalence of chronic non-communicable diseases [7][8][9][10][11][12][13] . The historical context of the contact of these peoples with the non-indigenous society, their geographic location along a border, the proximity and ease of access between villages and urban centers -something common in Mato Grosso do Sul's villages -, associated with social problems such as unemployment, violence, exploitation and sexual abuse 14-16 indicate a situation of great vulnerability that may help to understand the concentration of two-thirds of the cases identified in this study between these two ethnic groups, especially in the villages belonging to the Dourados, Amambai and Iguatemi Health Base Poles.
Furthermore, the mobility between villages -even outside the Brazilian territory since the Guarani and Kaiowá territories extend to Paraguay 8,9 -hinders the regular access to health services and diagnosis. Situations of intense inter-ethnic contact and marginalization in the access to health services are also risk factors for the circulation and spread of HIV. Socioeconomic, environmental, and cultural determinants interfere with the intensity of disease spread, making prevention and control activities complex, as evidenced in a study on the introduction of AIDS in the Xokleng people of Santa Catarina 17 .
In this study, the population with the greatest vulnerability to the spread of HIV and AIDS were young people, especially women with low schooling level. Unfavorable socioeconomic conditions are associated with greater exposure to risk factors. Schooling is a social status marker and has effects on the access to information and knowledge about disease prevention methods and its treatment 18,19 , including for indigenous peoples. Studies have shown that schooling is also directly associated with the survival of patients with HIV and AIDS, having a protective effect against most causes of death 20,21 .
The findings of this study differ from those observed among the indigenous peoples of Peru and Bolivia, whose HIV infection/AIDS ratio among men and women was 3:1 and 5:3, respectively 19  On the other hand, the limitations of the diagnosis of HIV infection in the male indigenous population may be associated with the low coverage of diagnostic tests, absence of men in the villages for work-related reasons, and reduced male demand for health services. The low diagnostic coverage favors the spread of the infection since ignorance about the disease leads to the continuation of sexual practices without prevention. The gradual inclusion of rapid HIV testing in villages as of 2012 may have contributed to the resumption in the growth of the HIV detection rate in the last two years of the study period since it led to increased diagnostic coverage, as observed in other DSEIs across Brazil 24 . The predominance of AIDS cases diagnosed in the 20 to 39 years and 40 to 60 years age groups was also found in the Brazilian scenario b and in a study conducted among the indigenous peoples of Bolivia and Peru, which found one in two cases among 20 and 39 years 19 .
The observation of mortality and fatality rates among the Kaiowá is relevant for the organization of health services since it indicates that this ethnic group has less access to antiretroviral therapy. The decrease in mortality and fatality rates in the Dourados Base in the second period suggests a greater structuring of health services in the villages, and articulation with specialized care services (SAE) for the treatment of indigenous individuals with AIDS. The differentiated behavior of the HIV and AIDS rates in each of the Health Base Pole shows the need for the separate evaluation of health services, given that these present different structuring forms to address the problem.
The main limitations of this study are related to the use of secondary data sources, which compromise the quality of information. The deficiency in the completeness of the variable indigenous race/skin color in Sinan and the underreporting of HIV and AIDS cases in indigenous peoples are factors that influence the number of identified cases and, consequently, information coverage.
This study used daily EMSI records as a data source to improve the coverage of HIV infection cases, since compulsory notification was only included in Sinan in 2005 for pregnant women g , and in 2014 for the general population h . Underreporting and incompleteness of the variable race/skin color data make it difficult to characterize health inequity situations, and to establish strategies for coping with such inequities 25,26 .
Despite these limitations, this study has brought benefits by addressing a disease that is not very well known in the epidemiology of indigenous peoples in Brazil, which impacts on the morbidity and mortality profile, and is related to the emergence of sexually transmitted infections and its association with social and cultural changes resulting from the process of interaction with the non-indigenous society.
The challenge of working with sexually transmitted diseases in an inter-ethnic context implies the need to comprehend sexuality values inherent in each indigenous group 27 and the articulation between indigenous and non-indigenous health systems 28 . Knowledge of logistical difficulties, population distribution and more vulnerable areas help to define strategies to expand the diagnostic coverage and facilitate the access to treatment 29 . In this sense, the training of human resources to work in intercultural contexts is fundamental to increase response capacity Sasi-SUS in disease prevention and treatment actions.
This study showed that the endemic character of HIV and AIDS -with prevalence of HIV infection in the southern region of the state -, may become epidemic in some years since there are cases in other indigenous areas of Mato Grosso do Sul. Among the crucial intervention measures to address the problem, we can cite the expansion of diagnostic coverage, the guarantee of access to treatment, and measures to prevent HIV transmission.