INTRODUCTION Currently, the epidemiological picture of AIDS in Brazil is marked by feminization, internalization, impoverishment and increased incidence in the young population. During adolescence, the period between 10 and 19 years of age, feminization is most marked ¹ . In Brazil, the prevalence rate of HIV infection in the young population shows an increasing trend and in the case of adolescence, the percentage of HIV infection becomes more significant if we consider that AIDS manifests itself on average six years after infection. Thus, it can be considered that a large part of the individuals who were notified in adulthood became infected in adolescence or early youth ^2,3,4 . The first case of AIDS among adolescents reported in the city of Recife occurred in 1985, and this was a male case, and only in 1987 was a female case diagnosed. In 2005, 20 years after the first reported case, the same number of cases were reported between the two sexes, demonstrating the feminization of AIDS at this stage of life ⁵ . Despite the proven national experience in controlling the AIDS epidemic, this problem continues to challenge public policies, especially when related to the young population. In Brazil, there are many studies on AIDS, however, few focus on adolescence. Thus, the objective of this research was to describe the epidemiological profile of adolescents living in Recife-PE, reported as HIV/AIDS cases between 2007 and 2015. METHODS A descriptive population-based study was carried out using the database of the AIDS Notifiable Diseases Information System (SINAN_AIDS) of the Municipal Health Department (SMS) of the City of Recife (PE), population data from the Brazilian Institute of Geography and Statistics (IBGE) up to 2012 and population projection data for the City of Recife for the years 2013 to 2015. The database for analysis was formed by combining the databases of child AIDS (under 13 years old) and adult AIDS (≥ 13 years old). The selected variables were classified into: a) Sociodemographic characteristics (age, sex, race/color, education, district of residence); b) Clinical and epidemiological characteristics (category of exposure, probable mode of transmission, CD4 T lymphocytes _< 350 cells/mm3 ^; criteria for defining an AIDS case, number of associated opportunistic diseases, case outcome). Data were analyzed using the Statistical Package for Social Sciences (SPSS) 20.0 according to the blocks of variables. Measures of central tendency were calculated for continuous variables and differences in proportions of categorical variables were compared using the t-test.Chi-square test , considering a p -value less than or equal to 0.05 (p≤0.05) as a statistically significant difference. Incidence and mortality indicators were also calculated by year of diagnosis and age group. The research project was approved by the research ethics committee under CAAE number: 55201016.6.0000.5207. RESULTS During the study period, 84 cases of AIDS were reported in adolescents aged 10 to 19 years living in Recife, PE. The average incidence rate for the period was 3.2/100,000 inhabitants, with 2014 presenting the highest incidence rate (6.2). Analyzing the incidence rates (CI) by age group, a decreasing trend was observed for adolescents aged 10 to 14 years and an increasing trend for those aged 15 to 19 years (Figure 1). The mean age of cases was 17.4 years (±SD 2.2 years) and the median was 18, with the age group of 15 to 19 years being the most prevalent (89.3%). Males and females had the same number of cases during the study period. However, a marked difference was observed between the ratios (%), being higher in 2010, when there were 500 female cases for every 100 male cases. From 2011 onwards, this feminization began to decline, reversing in 2013 (Figure 2). Comparing the proportions of covariates by sex (Table 1), a statistically significant difference in proportion was observed only for the categorical variables of exposure and mode of transmission. When the variables were analyzed, removing the ignored information and comparing the relative values ​​of the covariates by sex, a higher prevalence was observed in the age group of 15 to 19 years in both sexes. However, females had a higher proportion of cases in the age group of 10 to 14 years (16.7 versus 4.8%), brown race/color (84.6 versus 63.9%) and illiterate/incomplete elementary education (63.2 versus 43.3%). Regarding the distribution by District of Residence, it was observed that although there was a higher concentration of cases in Health Districts I, IV and V, there was no statistically significant difference in proportion (p≥ 0.05). Table 2 presents the clinical and epidemiological characteristics of the cases. Sexual transmission prevailed in both sexes, although it is noteworthy that 5.6% of the adolescents reported that transmission was through injectable drug use. Regarding the category of exposure, the majority of the adolescents are heterosexual (77.5%) while 48.4% are homosexual. It was observed that the CD4 ⁺ T lymphocyte count lower than 350 cells/mm3 ^was the highest percentage for both sexes, being 76.2% and 66.7% for females and males, respectively. Regarding opportunistic diseases, in most cases there were no records of diseases for either sex. However, among those who presented diseases at diagnosis (47.6%), 79.0% of the adolescents and 61.9% of the adolescents presented more than two opportunistic diseases. Regarding the evolution, 13 (15.5%) cases were registered as deaths, of which four (30.8%) were female and 9 (69.2%) were male. The average mortality rate per 100,000 inhabitants for the period studied was 0.9/100,000 inhabitants. Regarding the mortality coefficient (MC), an increase was observed over the years, going from 0.8 in 2007 to 1.6 in 2014 (Figure 3). Observing by age group, the years with the highest values ​​were 2011 and 2012 among individuals aged 10 to 14 years, and 2011 among those aged 15 to 9 years (Figure 3). DISCUSSION In this study, the population of adolescents aged 10 to 19 years reported as AIDS cases in Recife between 2007 and 2015 was characterized by equal proportions between the sexes. This finding points to the feminization characteristic that the epidemic has reached. Compared to the beginning of the epidemic, when people living with the disease were predominantly male, 30 years later, the incidence between the sexes reached an equal proportion ⁶ . When the incidence rates were analyzed by age group, it was observed that the highest rates were among adolescents aged 15 to 19 years, with an increasing trend. The higher incidence in this age group can be justified by the fact that the disease manifests itself on average six years after infection, for those who were infected in childhood. In addition, it is assumed that it is in this age group that most adolescents begin their sexual life, being more likely to acquire sexually transmitted diseases. In line with the incidence, mortality was also higher among individuals in this age group ² . Although the study period was less than 10 years, which may make trend analysis difficult, the incidence rates calculated per year may demonstrate the trend of an increase in the epidemic among adolescents, which has also been proven in other studies, reaching its peak in 2014, with a CI of 6.2 per 100,000 inhabitants. Regarding mortality, the CM showed a significant increase from 2011 onwards, remaining fluctuating with almost constant values ​​until 2015. Regarding the decline in the feminization of AIDS found from 2013 onwards, this result may be a reflection of collective actions that were articulated by UNAIDS with 10 goals to be achieved by 2015. Among these goals are: eliminating gender inequalities and gender-based abuse and violence, and increasing the capacity of women and girls to protect themselves from AIDS. With the reduction in the feminization of the epidemic, it is also possible to achieve a reduction in vertical transmission of AIDS, bringing benefits to the entire population in general ⁷ . Still regarding the sociodemographic characteristics of adolescents, race/color, education and age group were shown to be predominant characteristics for both sexes. The majority are brown, with incomplete elementary education and are between 15 and 19 years old. It is possible to consider that there is a low level of education in the majority of this population, since in this age group, adolescents are expected to have completed high school or be attending high school. Thus, studies point to the influence of the social context on vulnerability to AIDS, since low education makes it difficult to access correct and reliable information about disease prevention, especially STD/AIDS ^8,9 . Regarding the Health Districts, although some concentrate people with greater purchasing power than others, when observing the general distribution of adolescents by SD, there is no significant discrepancy between them, reflecting the heterogeneity of social classes. Thus, an epidemic that was previously characterized by affecting rich people, today appears to affect people regardless of social class ¹⁰ . Regarding the category of exposure, the highest percentage of cases were among heterosexual women and homosexual men, converging with the history of the epidemic. This data is historically compatible, since initially the disease affected more homosexual men, and from the sexual contact of these men with bisexuals, heterosexual women began to be infected with AIDS. This difference in the predominant category between the sexes is confirmed by other studies conducted with adolescents, reflecting that the tendency for female submission and difficulty in negotiating safe sex is still striking ⁹ . Despite the efforts of the Ministry of Health to reduce sexual transmission of the AIDS virus, it is prevalent in this age group, especially in the female population. This may be due to the difficulty in adhering to safe sex practices in this age group, as well as the challenges adolescents face in negotiating condom use with their partners. Comparing another study conducted with adolescents in Rio Grande do Sul in 2008, 61.54% had a vertical form of transmission. This differs from the present study, where sexual transmission accounted for the highest percentage of cases and vertical transmission accounted for only 7.1% in girls and 4.8% in boys. This finding may be related to advances in prophylactic drug technology during pregnancy and childbirth ¹¹ . The TCD4+ lymphocyte count was lower than 350 cells/mm3 in most adolescents. This may be related to the delay in diagnosing the disease and the consequent delay in starting treatment, thus aggravating the infection. Thus, it is observed that there is a need for timely/early diagnosis through prevention actions and expanded testing. Furthermore, it can be questioned whether there are flaws in the health care network, since this population does not live in the interior and there are several specialized HIV/AIDS services in Recife. Although most adolescents were notified with a CD4+ T lymphocyte count of less than 350 cells/mm3, more than half of the girls and half of the boys did not have opportunistic infections when notified, followed by a higher percentage of 1 to 5 opportunistic diseases, reaffirming that the evolution of the epidemic was not only modified by antiretroviral therapy (ART), but the clinical management of opportunistic infections is also a result of changes in the pattern of the epidemic ¹¹ . The increase in mortality found in this study contradicts the statement presented in 2013 by the Brazilian Ministry of Health, when it said that since 2007 there has been an increase in the survival of people with AIDS in Brazil. This result demonstrates that, despite the efforts made in Brazil and worldwide to reduce the prevalence of HIV infections and increase the survival of people living with AIDS, there are still more vulnerable populations that continue to become infected, living with the disease without it being controlled in a way that prevents death ^2,12 . CONCLUSION This study presented many limitations regarding the incompleteness of some variables in the SINAN/AIDS database of the city of Recife, which hindered a more accurate analysis of the epidemic among adolescents. The delay in reporting, demonstrated by the number of cases reported only at the time of death, results in an underestimated epidemiological picture that can lead to erroneous interpretations about the evolution of the disease. We emphasize the importance of investments in reporting cases and increasing testing actions, through training and awareness about the importance of filling out the notification form. These actions are necessary to achieve greater completeness and decrease the time between diagnosis and reporting, and consequently, better planning of health actions. This research affirms the importance of attention to adolescent health, since the results indicated clinical epidemiological data that could have been avoided or obtained more positive results, such as, for example, avoiding sexual transmission of HIV in this age group. This age group has many ways to obtain information about safe sex, such as increased access to the internet, open dialogue between families, and schools, for example. However, the results of this research show that this ease of access to information does not minimize the risk of illness. Therefore, it is essential to create welcoming means for this age group to seek information and support from health services. In this context, the School Health Program (PSE) can help establish a bond with the target audience. NOTE OF ACKNOWLEDGMENTS It is essential to thank everyone involved in preparing this article, especially Maria Goretti de Godoy and the entire team at the DS/AIDS and Viral Hepatitis Coordination Office of the Recife City Hall, for their receptiveness, goodwill, and trust in providing the data used in preparing this work.