Youth and Reproduction : Demograhic , behavioral and reproductive profi les in the PNDS-2006

MÉTODOS: Estudo populacional transversal com representatividade nacional sobre o comportamento sexual, contraceptivo e reprodutivo de 2.991 mulheres de 15 a 20 anos na Pesquisa Nacional de Demografi a e Saúde da Criança e da Mulher 2006. As jovens foram classifi cadas em três grupos: iniciaram a vida sexual e engravidaram antes dos 20 anos (grupo A); iniciaram a vida sexual e não engravidaram antes dos 20 (grupo B) e não iniciaram a vida sexual (grupo C). Mulheres de até 25 anos foram consideradas para o estudo das razões da gravidez e de suas implicações na vida. As análises estatísticas consideraram os pesos e o planejamento amostral complexo. A associação entre duas variáveis categóricas foi avaliada pelo teste tipo qui-quadrado. Quanto às comportamentais, utilizou-se modelo linear global.


INTRODUCTION
The fertility rate for the age group of 15 to 19 years, which increased 25% between 1991 and 2001 in Brazil, b began declining in 2000.a Nonetheless, the share of overall fertility has increased for youth in Brazil.b The contribution of fertility among women age 15 to 24 years to the total fertility rate (during the entire reproductive period) went from 34% in 1980 a to 53% in 2006.c This increase occurred primarily due to an increased relative participation in fertility among 15 to 19 year olds, which increased from 9% to 23% during the period.
Literature about sexual and reproductive behavioral patterns of adolescents and youth in Brazil, although rich and lengthy, 6,8 lack national level studies and/ or household surveys.In addition to the National Demographic Survey on the Health of Children and Women (Pesquisa Nacional de Demografi a e Saúde da Criança e da Mulher, PNDS) of 2006, performed by the Brazilian Center of Analysis and Planning (Centro Brasileiro de Análise e Planejamento, CEBRAP), there are the 1996 PNDS, performed by the non-governmental organization Family Well-Being (Bemfam) and two nationally representative editions of the study on Sexual Behavior and Perceptions of the Brazilian Population about HIV/AIDS, carried out in 1998 d and 2005. 2 Studies concerning reproduction among women age 20 years and younger, which represents about 16% of all women of reproductive age, e deserve the attention of researchers in diverse fi elds, especially in the areas of sexuality, gender and reproductive health.
The present article sought to analyze sociodemographic characteristics and the sexual and reproductive behavior of young women.

METHODS
A cross-sectional household survey was performed with 2,991 women of age from 15 to 20 years.The data are from the database of the 2006 PNDS.
The 2006 PNDS is a nationally representative household study with a complex probabilistic sampling design.Sampling units are selected in two stages: primary units, composed of census sectors, and secondary units, composed of household units.The study universe consisted of specific households located in common or "not special" census sectors (including favelas), selected from ten independent sample strata, consisting of a combination of the fi ve macro regions of Brazil and their urban and rural areas.The 2006 PNDS describes the profi le of the female population at reproductive age and of children less than fi ve years old in Brazil, as well as identifi es changes that occurred in the health and nutritional situations of these two population groups during the previous ten years.The database is available over the internet.f Questionnaires were administered face to face, after taking a complex probabilistic sample that is nationally representative for the fi ve Brazilian regions and for urban and rural settings.g The study universe included 14,617 households and 15,575 women from age 15 to 49 years.Each woman was asked about her sexual and reproductive history.
Youth were classifi ed in three groups: Group A began sexual activity and became pregnant before 20 years (n = 741); Group B began sexual activity and did not become pregnant before age 20 (n = 928); and Group C did not begin sexual activitybefore age 20 (n = 1,314).Eight women could not be classifi ed into these groups due to missing information about sexual initiation.
Socioeconomic and demographic variables included: region (North; Northeast; Central-west; Southeast; South), location of residence (urban; rural), skin color (white, black and other), marital status (married/partnership; no), years of study (0 to 4; 5 to 8; 9 or more), economic class (A + B; C; D + E) and health insurance or health plan.
Age (in years) at fi rst sexual relation (groups A and B), at fi rst pregnancy (group A) and at fi rst partnership (living with your fi rst partner/husband, groups A and B) were the behavioral variables.
The study analyzed knowledge of contraceptive methods, sources and type of contraception utilized at fi rst sexual relation and at time of interview, motivations for not currently using contraceptives, and knowledge of reproductive physiology.
Women age 25 years or younger (n = 1,771) were included in the study about reasons for the pregnancy and its implications on life.Positive and negative aspects of changes to one's love life, conjugal life and social life were addressed.
Statistical analysis was performed considering the weights and the complex sample design.g Nonetheless, the number of cases were presented without weights in order to provide the actual numbers on which the analysis was based.
The association between the dependent and independent variables was evaluated by a chi-square test adjusted for the sample design.The evaluation of behavioral variables was done with a linear regression model incorporating information on survey weights and the sample design.Statistical signifi cance was assessed at p < 0.05.
Analyses were performed using the software programs Stata v.9 and/or SPSS v.14.

RESULTS
Of the total sample of young women from age 15 to 20 years, 44.2% had not had sex, 31.1% had sex but did not become pregnant, and 24.8% became pregnant before 20 years.
Group A consisted primarily of young black women, of lower educational attainment and less income (Table 1).
The prevalence of not initiating sex before age 20 varied from 47.4% in the Northeast to 29.9% in the North.The highest percentage of women that did not become pregnant lived in the South region (42.7%), and the highest percentage of women that became pregnant before age 20 was in the North (37.6%).
Although the prevalence of sexual initiation before age 20 years was the same between whites and blacks (41%), black youth had a statistically signifi cant greater proportion of pregnancy before age 20.
Greater education was associated with an increased proportion of women not having sex before age 20 years (p < 0.001).The proportion of youth that became pregnant before age 20 years was highest (60.6%) for those with zero to four years of education, decreasing to 17.3% for those with nine or more years of education.
Greater poverty was associated with a greater proportion of sexual initiation before age 20 years (65.1%) for economic classes D and E and 41% for classes A and B (p < 0.001).The same association occurred for pregnancy before age 20 years.Skin color did not interact with the association of education and of economic class to sexual and reproductive behavior (Tables 2 and 3).Women that did not become pregnant before age 20 years (Group B) initiated sexual activity later at 15.8 years, a statistically signifi cant greater mean (p < 0.001) than the 14.9 years for women that became pregnant before age 20 years (Group A).
The reasons presented by the young women of group C for not having initiated sexual activity were a lack of desire (33.6%), intention to marry as a virgin (29.9%) and not having met a partner (25.2%).Fear of pregnancy (9.2%) and fear of AIDS (2.1%) were among their worries.
Knowledge of contraceptive methods was nearly universal among the groups, just as for all women of reproductive age.The pill and the condom were the most known methods among all groups.
In group A, the level of knowledge about the diaphragm, creams-suppositories and the rhythm method reached lower levels when compared to groups B and C. Youth without sexual experience presented less knowledge about withdrawal (48.5%) than women who had sexual relations (78.0%).This did not differ between groups A and B (p = 0.840), and both groups differed from group C (p < 0.001).
The day-after pill, although it was infrequently reported in spontaneous responses, was well diffused among youth when questioned, with greater knowledge among those who never became pregnant (86.0%).
Youth that reported not using contraceptives showed a low degree of knowledge in regards to locations for obtaining them (approximately 38%), without statistically signifi cant differences between the three groups.
The Public Health System (Sistema Único de Saúde -SUS) was the most common location (approximately 90%) followed by pharmacies and health plans.
For current users of contraceptives, pharmacies were the most utilized source by the youth of both groups (56.4%, group A; and 77.3%, group B), followed by the SUS (39.6%, group A; 17.3%, group B).The portion of women that used SUS to obtain contraceptives was relatively greater in group A, which corresponded to poorer youth, with a lower level of education and less access to health plans.
The youth of group A presented a lower percentage of contraceptive-uptake at first sexual relation, with a statistically signifi cant difference of 18.5% in relation to group B (p < 0.001) (Table 4).The same situation occurred in regards to condom use: 61.7% of women that did not become pregnant used a condom, a signifi cantly higher percentage than in group A (49.8%; p = 0.013).The use of contraceptives at least once was universal.There were statistically signifi cant differences between the groups in regards to current contraceptive usage (72.2% of group B and 56.7% of group A).The same situation occurred for the use of any contraceptive or a condom during the fi rst sexual relation.
Among youth of group A, 18.2% reported the use of the day-after pill at least once in their life, a statistically signifi cant lower percentage (p = 0.036) than group B (29.7%).The most common reasons for non-use of contraceptive methods were: "have infrequent/few sexual relations" and "not being married or with a partner", for both groups.Almost 20% of youth from group A did not use contraceptives due to breastfeeding/recent pregnancy, which could expose them to a new pregnancy.
Among the women, 77.3% of group B and 63.7% of group A affi rmed that there is a period between menstruations when a woman can become pregnant.This knowledge was signifi cantly lower (p < 0.001) among those that never had sexual relations (55%) (data not shown).
A small percentage of correct responses were observed in regards to the likelihood of becoming pregnant during the fertile period.No difference was found between these women and women that did not become pregnant at fi rst sexual relation, when considering education, income and skin color.The Northeast had the highest percentage of pregnancy at fi rst sexual relation (34.0%), a prevalence that was also higher in rural areas.
Of women that became pregnant before age 20, 18% did not marry before this age.Of those that entered partnerships (82%), the union happened after the fi rst pregnancy in 18.5%, and 35.5% married before becoming pregnant.No sociodemographic variable was significantly associated with first pregnancy occurring before or after the fi rst partnership.The age at fi rst pregnancy and fi rst partnership coincided for the remaining 46%.Since detailed information on age was not available in months or days, it was not possible to identify if the pregnancy happened before or after the fi rst partnership for the above mentioned 54%, in accordance with the criteria used.
The average age at fi rst partnership was 15.9 years for group A, younger than the 16.5 years for women in group B.
Women entered partnerships earlier in rural areas, and greater education, higher economic standing and white skin color were associated with later partnerships, per univariate analysis.Earlier initiation of sexual activity was associated with earlier partnerships.
There were various reasons that best explained pregnancy before age 20 years among the youth of age up to 25 years (Table 5).
Although almost all women of 25 years or younger, reported current familiarity with contraceptive methods, 71% reported that method failure or lack of knowledge about contraception was a reason for pregnancy before age 20 years.There was not a signifi cant association between this variable and the sociodemographic characteristics.
Desire to be a parent was as high as 52.6% with significant differences according to economic class.Lower economic status and lower education, were associated with higher percentages of women reporting the desire to be a mother/parent as a reason that best explained the occurrence of the pregnancy.
Conjugal unions before age 20 years were the reason for pregnancy for 51.1% of all youth and an even greater percentage of youth with a low level of education (63.3%).
Desire to marry was one of the reasons for pregnancy before age 20 years among 38.1% of youth age less than 25 years.This motive was infl uenced by level of education and economic status, with signifi cantly more infl uence among youth with four or less years of education (42.7%) and from economic classes D and E (41.9%).A higher percentage of women, who were not at school at the time of interview, reported a desire to marry as a reason (40.9%), when compared to those who continued their studies (26.9%; p = 0.005).
Difficulty to access contraceptive methods was reported by 26.4% of women, and greater poverty and lower education were associated with a more diffi culty in access.
A desire to leave their parents' home was reported by 23.4% of youth.At lower levels of education and economic status, greater percentages of youth reported this reason.
The percentage of youth that responded "no other option" (16.4%) was low and without association to the variables analyzed.
The perception of youth in regards to implications that resulted pregnancy upon their love lives and self-esteem was positive, independent of economic class or level of education.In regards to subjective aspects (selfesteem), 96.2% reported thinking of positive implications.Nonetheless, from the perspective of love and conjugal union, there was greater controversy; while 62.6% considered pregnancy as positive, 23.8% saw little or no positive implications (Table 6).
In regards to the totality of the situations considered as negative ("family rejection", "spousal abandonment", "life became more diffi cult", "lost my friends" and "more diffi cult to date"), herein referred to as the social dimensions of life, there was a widespread perception of little negative impacts from pregnancy.Although the implications that were considered as negative only reached a low percentage (7.6%), the perception of negative implications was signifi cantly higher among poorer (9.9%) and less educated (10.6%) women.
of current contraceptive use and contraceptive use at fi rst sexual intercourse, which demonstrates more risky behavior compared to women of higher social standing.Contraceptive use is one explanation for why women of group B did not become pregnant.
The results on sources for contraception show that women of lower social status have a greater dependency on the Public Health System (SUS) in order to meet their reproductive rights.In the same manner, the signifi cant difference encountered between the economic groups, for the use of the day-after pill, may indicate a greater diffi culty in access to emergency contraception among women of group A.
There was a signifi cant association between pregnancy before age 20 years and increased poverty and lower education.On the other hand, not all youth that began sexual activity became pregnant.Among those that became pregnant, the average age of sexual initiation was younger than those that did not become pregnant.
There were various reasons for pregnancy before age 20 years.The majority of youth became pregnant in part due to lack of knowledge about contraception or contraceptive failure, independent of education and economic status.This result can be understood within the context of sexual experimentation, 4,9 which involves negotiating gender norms in addition to the diffi culty in managing the appropriate use of contraceptive methods.
Compounding this is a widespread lack of knowledge of reproductive physiology, as was also identifi ed by other studies. 5,7This issue is a shortcoming of both research and interventions, which merits attention.
The occurrence of pregnancy is often considered an absence of a life plan.Access to better education, better living conditions and greater opportunity would favor increased options during youth trajectories. 8,10,13  the absence of these structural and infl uencing conditions, pregnancy, although unplanned, constitutes a life plan and not an absence of planning, as found in the present study.
Among the positive and negative implications from pregnancy, women perceived more social and personal advantages from this reproductive decision.International studies show similar results.Despite great advances, it is necessary to contemplate actions that address the diversity of Brazilian youth.In order to better understand the dynamic of youth trajectories regarding the transition to adulthood and its perceived benefi ts, interventions should consider the economic, social and cultural determinants for preventing pregnancy, sexually transmitted diseases and HIV. 9 Sexual education programs in public schools have been minimally implemented. 12In addition to structural and cultural factors, institutional factors that infl uence reproductive behavior should be considered.b "Very positive" when the women agreed with at least 3 of the following items: "more respected as a mother", "relationship with husband improved", "married and made a home"; "Somewhat positive" when agreed with 2 items; and "Little or no positive implication" when agreed with only 1 or no items.c "Very negative" when the women agreed with at least four of the following items: family rejection, partner abandonment, life became more diffi cult, lost my friends and became more diffi cult to date; "Somewhat negative" when agreed with 3 items; and "Little or no negative implication" when agreed with 2, 1 or no items.d "Positive" when the woman agrees with at least one of the following items: feel better about myself and have reason to live; and "Negative" when they did not agree with either item.

a
Berquó E, Cavenaghi S. Increasing adolescent and youth fertility in Brazil: a new trend or a onetime event?In: Annual Meeting of the Population Association of America; 2005 Mar 30 to Apr 2, Philadelphia, USA; 2005.b Berquó E. [Increasing youth fertility].In: Encontro Estadual de Políticas Públicas da Juventude, São Paulo: Associação de Apoio ao Programa de Capacitação Solidária; 2004.p.185.
The 2006 PNDS was approved by the Research Ethics Committee of the STD-AIDS Reference Center of the Secretary of Health of São Paulo State (3 Oct 2005, Protocol number 029/05).

Table 1 .
Sociodemographic variables of women age 15 to 20 years, according to groups A, B and C. Brazil, 2006.

Table 2 .
Education and skin color of women age 15 to 20 years, according to groups A, B and C. Brazil, 2006.

Table 3 .
Economic class and skin color of women age 15 to 20 years, according to groups A, B and C. Brazil, 2006.

Table 4 .
Use of contraceptive methods at first sexual intercourse, for groups A and B. Brazil, 2006.
The p values correspond to chi-square tests adjusted for study design, comparing the percentage of group A and B for each line.Source: PNDS 2006

Table 6 .
Percentage of women according to the implications that result after they became pregnant before age 20 years.a Responses valide for women age 25 years or less at time of interview and who had live born children before age 20 years. a