Introduction

Malaria is a worldwide health problem, with the majority of cases occurring on the African continent (93%), followed by the Southeast Asian region (3.4%) and the Eastern Mediterranean region (2.1%). In 2018, it is estimated that there were 228 million cases of malaria worldwide, in comparison to 251 million cases in 2010, and 231 million cases in 2017. The incidence rate of malaria decreased globally between 2010 and 2018, from 71 to 57 cases per thousand at risk inhabitants [1]. Despite being avoidable and treatable, malaria continues to have a great impact on human health and its subsistence [1].

In the Americas, despite having a lower prevalence when compared to global levels, malaria is a severe problem. Three countries (Brazil, Colombia and Venezuela) account for 80% of all estimated cases in the Americas, which 51% of the cases occurred in Venezuela, followed by Brazil with 23% of the cases in 2018 [1]. In Brazil, the majority of cases (99%) are concentrated in the Amazon region, which is linked to environmental factors that favor the occurrence and transmission of malaria such as high temperatures, high precipitation and humidity, low altitude and its extensive network of rivers [2, 3]. Furthermore, environmental changes due to anthropic activity, especially Forest clearing, cause great impact on the malaria transmission. Such alterations favor vector breeding sites that, when combined with the population's lifestyle and its demographic and socioeconomic characteristics, impact on the occurrence of malaria. In this context, the indigenous population may be more vulnerable to malaria, due as much to the rural location of their homes as to their lifestyle [4].

The state of Amazonas reported more than 82,000 new cases of malaria in 2017 [5]. This state also has the highest proportional indigenous population in Brazil, as it has 183,514 indigenous inhabitants, which corresponds to 22.4% of the total of the country [6]. Although there are publications about malaria in the Amazon, information on malaria in indigenous populations is still poorly explored and reported in a generic manner [7–9]. Therefore, studies on the association between the occurrence of malaria and indigenous people, as well as the magnitude of the disease in this population, are necessary. The aim of this study was to the aim of this study was to identify factors associated with malaria in indigenous people, in a scenario of high malaria incidence in the state of Amazonas, Brazil, from 2007 to 2016.

Materials and methods

Results

Malaria cases were predominant in males in all categories, but in non indigenous people and missing race data this proportion was higher. In the indigenous category there was a predominance in the age group between 1 and 10 years of age, whose proportion decreases as age increases, while in non-indigenous and missing races groups the predominance occurred in the age range between 20 and <40 years of age. The low level of education was similar in both indigenous and non-indigenous groups, with higher proportions in individuals with only 1 to 4 years of schooling (Table 1).

Download:

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

Table 1. Characterization of malaria cases by location of probable infection, in indigenous patients (village or self-reported indigenous), non-indigenous and missing data on race.

Amazonas state, Brazil, 2007–2016.

https://doi.org/10.1371/journal.pone.0240741.t001

The passive case detection was more prevalent in all race categories. Parasitemia <+/2 (less than one parasite per 100 thick film fields) showed a higher percentage in the non-indigenous group; while parasitemia of ++ (11 to 100 parasites per 100 thick film fields) was higher among the indigenous group and in the cases where race was unspecified. In all racial groups, the main type of plasmodium was P. vivax, the predominant form of malaria in the region (Table 1).

In all categories of race, there is a gradual decrease in the number of malaria cases as increases the number of days between the onset of symptoms and diagnosis. For the amount of time elapsed between the first symptoms and the beginning of treatment, the longest period observed in all categories was up to 2 days (Table 1).

Regarding timeliness of treatment of malaria, the cases registered in all races categories (indigenous, non-indigenous and missing data groups) presented timely treatment longer than 48 hours: indigenous (55.7%), non-indigenous (55.0%) and those without data on race (50.3%) (Table 1).

The result of logistic regression analysis in order to estimate the factors associated with indigenous malaria with the respective values of crude OR and adjusted OR is presented in Table 2. In this analysis, we excluded race missing data cases. Although we can see in the Table 1, this category is similar to non indigenous.

Download:

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

Table 2. Factors associated with malaria (crude and adjusted analysis), according to indigenous patients (village or self-reported indigenous) or non-indigenous.

Amazonas state, Brazil, 2007–2016.

https://doi.org/10.1371/journal.pone.0240741.t002

In the multivariate analysis, the following variables were associated to indigenous: male sex (OR = 0.83; CI 95%: 0.82–0.84); patients younger than 1 year in indigenous populations (OR = 1.91; CI 95%: 1.80–2.03), whose OR decreased as age increased, at least up to 40 years; education has greater magnitude among illiterates (OR = 1.31; CI 95%: 1.28–1.35), with reduction of OR as the education level increased; active case surveillance was strongly associated to indigenous malaria (OR = 1.73; CI 95%:1.70–1.75); the highest parasitemia (++++, i.e. >10 parasites 2.73 per one thick film field) is associated with malaria in the indigenous population (OR = 2.10; CI 95%:1.53–2.88); P. vivax, mixed infections (up to twice as many cases in indigenous patients) (OR = 2.42; CI 95%:2.2–2.66) and other, i.e., infections caused by P. malariae and P. ovale were 4 times more common in cases of malaria in the indigenous group (OR = 4.82; CI 95%: 2.49–9.35); timeliness of treatment, longer than 48 hours, appears as a malaria protective factor in indigenous (OR = 0.98; CI 95%: 0.97–1.00) (Table 2).

Regarding malaria mortality in the period from 2007 to 2016, 109 deaths were reported (Table 3). Of these, two notifications were excluded due to lack of information or errors when filling in the forms. Considering indigenous groups, there were 34 deaths, 13 female and 21 male. Higher mortality levels were observed in those older than 60 years (32.3%). Regarding the non-indigenous group, there were 73 deaths, the highest number of deaths among female patients, and the most prevalent age group was between 20 and <40 years of age.

Download:

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

Table 3. Distribution of deaths caused by malaria, in indigenous and non-indigenous populations, Amazonas state, 2007 to 2016.

https://doi.org/10.1371/journal.pone.0240741.t003

Both categories (Indigenous and Non-indigenous) had underlying cause of death—ICD (International Statistical Classification of Health-Related Diseases and Problems) of death reported showing a greater proportion of P. vivax malaria.

Discussion

In the state of Amazonas, malaria is a disease that predominates in rural populations and reflects social and economic aspects that are accentuated in indigenous populations. Here the authors shed light on the factors associated with the occurrence of malaria in indigenous populations and cases that occurred in indigenous villages, as well as the description of the profile of deaths recorded by malaria in these populations in the state of Amazonas, by analyzing a period of 10 years (2007 to 2016).

The predominance of malaria in indigenous males is similar to the results found in other studies conducted in the Amazon Region [2, 11–14], which justify such predominance based on indigenous behavior and a lifestyle based on extractivism and family farming, which is predominantly performed by men who consequently are more exposed to contact with the vector and the parasite [15, 16].

The highest proportion of cases among children up to 10 years old in indigenous peoples found in this study is equivalent to historical studies in the Amazon [17, 18]. In a riverine community in Portochuelo (Rondônia), a higher prevalence of malaria was found in children under 16 years of age [19]. Similar results were found in a study with the Yanomami tribe, on the border between Brazil and Venezuela, where children under 16 were the most affected by malaria [20]. In other studies, results were also found in younger age groups (children under 14 years old), such as schoolchildren and preschoolers [11, 21, 22].

The malaria treatment guideline envisages immediate malaria treatment (timeliness of treatment) to be provided to all patients with positive slide results or positive rapid test results for both symptomatic and asymptomatic cases [23]. This fact may be closely related to the high proportion of cases with time of up to 48 hours between onset of symptoms and time of diagnosis and/or treatment time of up to 48 hours in indigenous people. A study conducted in the state of Amazonas, with a non-indigenous population, showed that most patients were diagnosed predominantly within three days of symptom onset (94.3%). The timeliness of treatment diagnosis and treatment helps to prevent hospitalizations and deaths, but also help to control disease transmission by preventing or reducing the onset of parasite (gametocyte) sexual stages in human hosts, which are the infectious forms in the vectors [24].

Regarding the Plasmodium species, the present study showed similar results to those performed in non-indigenous populations, if we consider that P. vivax account for majority of the cases [25, 26]. An epidemiological malaria survey found prevalence of P. vivax cases, followed by P. falciparum and mixed infection cases (P. vivax + P. falciparum) [27]. It is necessary to remark that we found highest prevalence of P. falciparum and mixed infection in indigenous population, which cause more impact in the indigenous in terms of severity fo the disease. Maybe this aspect can explain that in indigenous we had more cases in the passive detection, which leads people to look for diagnosis faster because of the symptoms.

In Brazil, the incidence of P. falciparum and P. vivax cases occurred at similar levels until 1988, after which the proportion of P. falciparum cases progressively decreased, while cases involving P. vivax increased and from then onwards became the predominant species responsible for over 90% of episodes of malaria [4, 5, 27]. It is important to note that there are biological differences between P. vivax and P. falciparum, which make P. vivax particularly challenging to control and eliminate in malaria endemic regions [28–30]. According to a study in Papua New Guinea, relapses may be the cause of about 80% of P. vivax cases [31]. A recent study in the state of Amazonas also suggests that relapses made a significant contribution to cases in this region [32].

The Timeliness of treatment found in this study can be interpreted as an effort to advance treatment initiation is due to the expansion of the laboratory network and the action of community health workers (CHA), which contributed to the decreasing trend observed in the percentage of malaria hospitalization of 3.3% in 1999 to 1.4% in 2009: a significant reduction of 79% [3].

The biggest obstacles to malaria treatment in indigenous populations is getting access to them due to the population dispersion and also the great distances involved as well as the local geography which both make the logistics of care more expensive. From the perspective of the diagnosis and treatment of malaria in indigenous peoples, the greatest difficulty is in reaching remote villages, due to the complexity of access which is only by river. The microscopy network does still not meet the total population coverage. The performance of well-trained and active indigenous health agents (IHAs) is a positive aspect for disease control [13].

In a study on factors associated with malaria treatment in the Brazilian Amazon, it was found that timeliness of treatment was associated with indigenous patients with low education (from zero to 5 years of schooling). These variables indicate vulnerable and highly dependent groups of the public Unified Health System (SUS) [23].

In endemic areas in Brazil, people are aware of the symptoms of malaria, and usually seek care early on in the clinical manifestations, which makes severe conditions uncommon. A survey conducted in the state of Acre, Brazil found that about 22% of patients sought care in less than 48 hours, while more than 70% of people looked for care less than four days from the onset of symptoms and most patients had malaria by P. vivax [33]. These are probably the explanations for finding only non-severe cases in this casuistry. This early search for diagnosis and treatment is also important to prevent people with protozoa in the peripheral circulation from acting as a source of infection for the disease transmitter, as proper treatment eliminates the blood parasite within a few days. This may shed light on the importance of timeliness of treatment and low mortality from malaria.

The adjusted multivariate analysis showed male is a protective factor for indigenous people, which differs from general population. This finding deserve investigation, since malaria transmission is related to activities that promote contact with vector, we do not know what occurs with indigenous.

Regarding age, the strength of the association of malaria in indigenous people is greater the smaller the age group. This fact corroborates another study in indigenous populations, in which the disease is associated with children [20]. On the other hand, it differs from other studies that detect a greater problem of malaria in the economically active population among non-indigenous people [23]. In the case of indigenous people, transmission may be intra- or peridomestic, where the vector is more likely to infect children, especially when they sleep.

The educational level must be interpreted with caution, since the indigenous populations, as they live in areas far from urban centers, may have more difficulty in accessing school. Perhaps this explains the greater chance of being a case among the illiterate.

The chance of malaria cases among indigenous people through the active case surveillance may result from the way in which indigenous health services are organized, in systematically promoting this active case surveillance. If we combine the interpretation with the timeliness of treatment, it is observed that even though the majority occurring within 48 hours, when it is more than 48h, it was a protective factor. Some explanation about the organization of malaria control actions in the Indigenous population can be found in a study on the municipality of São Gabriel da Cachoeira, which has the most indigenous people concentration in state of Amazonas [34].

The level of parasitemia can be an indicator of the severity of the disease among indigenous people, with greater strength of association in cases of highest parasitemia (++++). Still on gravity, the parasitic form has a stronger association for mixed infection (P. falciparum + P.vivax) and another parasitic form (P. ovale, P. falciparum + P. malariae or P. malariae), which reveals that indigenous people have particularities to be considered, since whereas in the state of Amazonas and in non-indigenous populations the predominance is of P. vivax.

The malaria mortality must be a rare event since the diagnosis and the treatment is freely available, and the health professionals are conscious of the importance in the case finding and management. However the mortality analysis reveals the unfavorable outcome, specially in male and older people. The underlying cause of death was caused by P. vivax with no complications in both population groups, indigenous and non indigenous. One problem with P. vivax infection is the possibility of causing unfavorable and unmeasured outcome [9, 35].

The limitations of this study are mainly related to the use of secondary data itself. For example, the lack of information on job occupation or other variable proxy to socioeconomic status did not allow to include in the analysis in order to find a better model adjustment. However, it is not possible to control report, missing data or typing errors, and also the underreporting malaria cases. The last cited can be a minor problem, since it is a compulsory notification disease, and the treatment is just made available after case notification. To tackle with race missing data, we proposed this method of race data imputation, which revealed to be useful to conduct this analysis. However it is known that, ideally, the 100% of completeness must be the goal of malaria control program.

Even considering the problems of malaria secondary data, we emphasize enhancing of the quality, its universal coverage in Brazil, the availability of great data amount and the possibility to work with indigenous data, with no need to go to field data collection, which in indigenous issues has other than costs implications.

Despite these limitations, we judge that not compromise the findings. They are useful in understanding of the malaria peculiarities in the indigenous population, since we found differences in the associated factors. We also remark the importance of the completeness of secondary data, mainly the variable race, because it can reveals cases profile, cases inequalities and contribute for malaria control.

Conclusions

The study findings presented here have identified the factors associated with malaria in indigenous populations and in other races. Among indigenous people, the disease predominates in men of a younger age and who have high parasitemia. Because indigenous inhabit places that with difficult access and with few health resources, the disease is more often detected via active case surveillance. The border municipalities which have indigenous populations presented higher than average percentage of malaria cases. Evidently there is a clear need to work on actions related to the planning and execution of health care for these locations.

Obviously neglected, indigenous people in the Amazon region are particularly affected by severe infectious diseases such as malaria, which adds an obstacle to both the quality of life of these populations and the advance of disease control and elimination in the region. The findings presented here clarify the malaria transmission profile in these populations, as well as their risk factors, generating evidence that supports decision-making by managers of the Malaria Control Programs in the Amazon Region.

Supporting information