Ayres explains that his formulation of happiness does not refer to the commonsense idea of the word. Rather, it is a regulating idea that guides our decisions, showing whether we are moving along in accord with our projects. It is, in his words, a counterintuitive and "assintotic" idea, insofar as we never manage to achieve the experience of complete happiness.

We argue here that care can be thought of as the construction of projects of the person which is expressed in everyday practices and which occurs in a framework of power relations between agents who occupy different social positions. We show, through ethnographic analysis, that for the subjects, care necessarily involves work focusing on the person, as Thomas (1993) argues. Our conceptualization of work as taking the form of the construction of projects in everyday practice may both be connected to Ayres' perspective and also held to be distinct from it. While Ayres restricts himself to examining care in a sphere delimited by restricted inter-subjectivity, in our research we show that it is in fact constructed culturally and socially within structured relations of power. The socio-historical aspects of care highlighted by Carvalho (1999) and other authors who work from a feminist perspective (Scavone, 2005), are also central to concept that is proposed here.

Projects of the person may be related to the multiple interests of caregivers occupying different positions within a social field, in the sense, with respect to this latter concept, ascribed by Bourdieu (1996, 1989). Such projects are not reducible just to a concern for practical success. Following Rabelo's (1999) concept of project, which derives from Schutz and Merleau-Ponty, we argue that projects involve more than simple discursive or mental constructions. Indeed, projects can have corporeal expression without necessarily having passed through a level of mental representations; what is more, several projects may coexist in the same situation. Based on Rabelo's contributions and upon anthropological discussion about the social construction of the person (see Bustamante, 2009), we argue that care (and with it the person) is always being built and rebuilt in this form - that is, as projects that indeed might not be spelled out discursively, in so many words.

This conceptual formulation draws upon a critique of a general tendency to universalize the meaning of care and, at the same time, proposes to widen the notion by showing that in fact care is built daily through diverse interactions, and not just out of a concern with happiness or well-being.

Based on this theoretical discussion, we argue that care is produced constantly in interactions between health professionals and users, and that it has different qualities which are related to the social positions of the interacting agents. The health professionals themselves understand that performing care depends on discourse - the act of giving guidance - and that it takes place in reference to planned routines. For them, care focuses on women seen as objects of these interventions. On the other hand the latter - the ‘users' - understand that care takes place in a spontaneous way, both with reference to corporeality and as part of responses to situations that arise unexpectedly. Such spontaneous and situational care is done by relatives of both genders (and not just by women). Moreover, users build into their practices a negotiation with the projects of the person that are proposed by the health professionals, whereas, by contrast, the professionals tend to keep to their original standpoint, without regard to the perspective brought to these encounters by the users.

With the research presented in this article we aim to contribute to a conceptual discussion that brings light to bear on day-to-day healthcare practices. This helps us to better understand the complexity of the situations that are part of the daily making of care and to identify those of its aspects that require further attention. This will help engender healthcare practices that are closer to the projects of the person - and of happiness - cherished by the users themselves.

Method

This analysis is part of a larger study on childcare in a low-income neighborhood of Salvador (Bustamante, 2009). The fieldwork, conducted between 2003 and 2006, involved contact with residents of ‘Prainha' (our fictitious name for the neighborhood studied) which is located in a district of Salvador known as the Subúrbio Ferroviário (Railroad Suburb). As part of our research, we also studied a variety of institutions serving children in the neighborhood.

In what follows we analyze data obtained through participant observation in these settings and through interviews conducted both with health professionals who worked in the local Family Healthcare centre and with some of the residents of Prainha for whom they cared. We focus, in particular, on the members of seven households. As Toren (1997) observes, participant observation is the method that best characterizes the ethnographic approach. Through the researcher being simultaneously a participant and an observer s/he is always asking questions about her own and other people's involvement in ordinary events. This attitude means that nothing that is said is considered irrelevant. As Toren points out, ethnographic analysis is not intended to be based on representative samples. Rather, the challenge is to know as much as possible about the people whose ideas and behavior are under view, and to be able to achieve this it is important to conduct in-depth interviews with some informants.

Ethnography is more than just a literary text in which the ethnographer tells the reader that she has, in fact, "been there" (Geertz, 1989). There are two criteria that we consider essential for thinking about the validity of ethnographic interpretations. On the one hand, as Jackson (1996) puts it, the credibility of discourse is not determined by seeming to allow the facts to speak for themselves, but rather by the way in which facts and data are organized into a narrative. On the other hand, as Pina Cabral (2005) emphasizes, interpretations should be constructed by starting out from a baseline - one where researchers and researched share a common world. He says: "The ethnographer, in his materiality, is co-existent with the subjects of the ethnographic study and it is only through the common world they share that he is able to make sense of his ethnographic observations." (Pina Cabral 2005:20) (Our translation)

As Wolf puts it, in this understanding ethnography is seen as a realistic work motivated "[….] by an urgent sense to place on record and testify to human experiences that ‘speak' to us, without flippancy, about things that matter "(Wolf (1983, p.xi), apud Jackson (1996, p.43)). Thus, ethnography is more than one type of writing; rather, it is the best way to understand and show how people from different groups live and how they relate to others.

As far as the process of analysis is concerned, this was ongoing throughout the study, taking place hand in hand with the writing process (Becker, 1994). The interviews and field notes were transcribed, read and organized into folders ordered chronologically. The first readings of the material were of a general nature, with the purpose of clarifying the main theme of the thesis, through identifying key points. A second type of reading followed, involving the identification of important issues through organizing excerpts in related sections, resulting in the creation of additional files. New readings of this selected material - and sometimes a return to the original notes - took place as the arguments of the thesis were built up. A deeper understanding of the body of data grew out of this process. Some of the most important findings came about after rereading previously scrutinized notes.

The research project that led to this article was approved by the Ethics Committee of the Public Health Institute (Instituto de Saúde Coletiva) of the Federal University of Bahia. Ethical procedures included the use of fictitious names in data dissemination. Full liberty was given to informants to decide on their participation, after being properly informed. Willing participants then signed the Consent Form.

Results and discussion

The Family Healthcare centre serving our research subjects was established in 2002. Each of the three Family Health Teams there is composed of a doctor, a nurse, nursing technicians, a dentist, dental technicians and community health workers. Each team serves approximately one thousand families. Just as in other Family Health centers, programmed actions in Prainha include family planning, prenatal care and childcare.

Day-to-day care of pregnant women and babies in Prainha occurs in the context of the changes that have affected Brazil in recent decades. There has been a dramatic reduction in the birth rate (Dalsgaard, 2006), as well as increasing reduction in infant mortality rates. As a result, the number of children per woman is lower than in previous decades. In Prainha, there are differences between mothers of young children and their mothers: Younger women tend to have fewer children and, as in other parts of Brazil, expressed the desire to have only two (Bemfam, 1997). Cases of infant death are rare in recent generations, unlike amongst the grandparents of small children, who tell of having had at least four children and often of having lost newborns or infants before their first birthday. On the other hand, in Prainha one frequently hears about abortions, which are usually induced with Cytotec (misoprostol), a drug that is available on the market for the treatment of stomach ulcers, the use of which has significantly decreased mortality rates due to insecure abortions in Brazil (Menezes, Aquino, 2009).

For reasons of space we concentrate here on analysis of care associated with the discovery of pregnancy and with the early prenatal period, as well as on care during the first months of the child's life, with particular emphasis on breastfeeding. We chose to focus on these topics because of their importance in the routine of health professionals and the lives of women and their families.

"To turn up pregnant" and the decision to continue the pregnancy

The health professionals interpret pregnancy among health center users as evidence of the failure of their "family planning" activities. Sometimes they lay the blame on their own work. For example, one day Dentist Ana[1] complained about a nurse's lack of commitment to her team, citing as an illustration her work in family planning (for which nurses are responsible), commenting that "if it were done right, no woman would "turn up pregnant" (aparecer grávida) here at all[2]". However, most healthcare workers blame the women themselves. Among the professionals the predominant opinion is that a woman who "turns up pregnant" does so because she failed to follow their professional advice.

Along these lines Nurse Rose opined that the women who use the center are "too lazy" to go to the clinic to seek contraceptive methods, or to bother to learn how to use them, and that often this is due to the passive way in which they lead their lives. In Rose's opinion a few "structured" families exist in which the couple actually plans to have children. In her view the possibility that a pregnancy be planned is associated with the existence of such a nuclear family.

"Family planning" is one of a nurse's responsibilities and so as part of her routine she should be available for consultations and should talk with women about contraceptive methods. On a day-to-day basis, however, the service offered at the center does not meet demand, especially since most women seek injectable contraceptives. A woman we interviewed informally in the waiting room explained this difficulty in obtaining contraceptives. As well as complaining about the scarcity of such contraceptive medication in the health center, she also grumbled that her husband was not allowed to withdraw it (when available) on occasions when she could not because she was at work. According to this informant, the health professionals are inflexible and, moreover, "when we turn up pregnant they say that we didn't follow their advice."

The situation faced by this informant is an illustration of the fact that health professionals' "family planning" actions are concentrated on women, as Schraiber (2005) found in her study of the everyday practices of the Family Health Program in Recife. She notes that professionals direct their actions at women seen mainly as mothers and caretakers of the family, who are expected to be available to spend many hours at health facility at the provider's convenience rather than their own.

The above discussion shows that both users and providers refer to pregnancy as something that "aparece" (happens, turns up, becomes apparent). This is consistent with the findings of the GRAVAD survey held in three Brazilian cities. Most of the mothers interviewed said they had become pregnant without planning (Aquino et al., 2003). However, residents of Prainha refer to "turning up pregnant" or becoming pregnant "without planning" in a different sense from the health professionals. The following excerpt from a conversation with Paula and Ed, a couple with two children, illustrates this. One day I asked[3]: "How did you used to imagine what it would be like to have children?" and Paula said, "No, actually I never thought about what it would be like, because it was not a planned thing. It happened, but when it did, I was sure I wanted it [...]"

The sisters Alicia and Lucineide both became pregnant "without planning", without a stable source of income, and both separated from their partners during the course of their pregnancies. However, each dealt differently with her circumstances. Alicia reported that she became pregnant because the couple did not do "family planning" during a very troubled period in which one of her children needed to be constantly taken to emergency services. She decided to "split up" after talking to her partner. He also decided to separate from her, firstly because he was away working in a country town, secondly because he had become involved with another woman. Alicia's mother thought it a huge "lack of responsibility" to have another child in these conditions, but nevertheless, accepted the arrival of her granddaughter. She helped when her daughter went into hospital, and provided care and support for Alicia's other children. During her pregnancy, Alicia also obtained aid from a Pentecostal church: Parishioners organized a baby shower to help her get clothes, diapers and other objects for the child.

Lucineide became pregnant with her third child at age forty, when she was already the mother of two teenagers. She says that she did not expect to get pregnant, but knew that if it did happen, she would not induce an abortion, because it was her responsibility to "evitar" (avoid) pregnancy - something that she had not done. She decided to "deixar" (literally "leave" in the sense of "allow to proceed") regardless of the opinion of her partner. Coincidentally, he also wanted to have a child.

These cases show that there are differences between individuals and also that partners, relatives and colleagues are important in the experience of pregnancy, not just in the decision to allow it to proceed, but also to deal with the emotional and material challenges it occasions. Nurse Rose's discourse contrasts with what we find in the daily life of women and their families. Relatives and sometimes partners participate in the decision to continue a pregnancy as well as in the care of pregnant women and also in the preparation for the arrival of the baby. What is more, this does not depend on the existence of a nuclear family structure.

The health professionals base their practices on the premise that the women are uninterested and lazy, and that they do not follow their advice. The occurrence of pregnancy is seen as evidence of these attitudes. The possibility that it is a wanted pregnancy is associated with the existence of a nuclear family, where a husband has the role of breadwinner. The insufficient supply of contraceptives is not discussed by the professionals as something that contributes to unplanned pregnancies. According to Nurse Rose, there is always some method available, sometimes only the male condom, but even so, women do not use it. (The relationship between poverty and a preference for longer-lasting methods of contraception, including surgical sterilization, has been discussed elsewhere (see Dalsgaard, 2006)).

Users make constant reference to spontaneity and physicality when talking of such matters. This can be adduced not only in their use of the expression "to turn up / be seen to be pregnant," (aparecer grávida) but also in the manner in which women like Paula speak about "displaying the belly" and "letting it grow". Pregnancy is an expected occurrence in women's lives, and especially so in the case of the first child. This may be a spontaneous strategy to constitute motherhood, since women know that, in most cases, health professionals do not view the open expression of a desire to have children as legitimate, because of their precarious living conditions.

We argue that "aparecer grávida" can be thought of as part of an implicit negotiation between users and professionals. Without leaving aside the problems of access to contraceptives, it is fair to say that this expression evokes something of an unconscious desire to have children. This point  is consonant with Menezes' (2006) discussion, based on the GRAVAD study conducted in three Brazilian capitals, which notes that it is common for women to say they did not actively want the pregnancy,  did not think about it, were not  trying to avoid it, and that the pregnancy just "happened.".

The care of newborn babies is also included in the officially prescribed routines set out in the Family Health Program. Giving advice on breastfeeding is of key importance to the program and we now turn to this topic.

Their "advice" and "our side of the story"

The following excerpt from the interview with Nurse Rose is illustrative of the health professional's point-of-view and its relation to the targets set by the Ministry of Health:

In the course of our research we confirmed Rose's observation that mothers often do not breastfeed, giving mingau to their children instead,  helped by older women. However, they try to keep this hidden from the health professionals. The first weeks of newborn Thaddeus provide a case in point. On the same day that his family proudly returned from the health center, after taking him for a check-up with the nurse and getting him vaccinated, I witnessed Dona Aurelina help her daughter-in-law Cristiane give mingau to Thaddeus. In Dona Aurelina's opinion ‘poor' women like Cristiane are not able to breastfeed a child, so in these cases mingau is the baby food of choice. When asked if she had talked about it with the professionals of the health centre, she said "No", and added: "One should not speak of this with the doctor, they do not understand our side of the story."

In our understanding this reluctance to reveal non-compliance on breastfeeding has to do with its association with good mothering. The following story involving Lucineide, her friend Claudia and the latter's husband Milton is illustrative:

In the midst of this conversation I realized that Claudia is married to Milton, a man whom I already knew. The first time I had talked to him, he was out walking with his two month old son in his arms. I was pregnant at the time and this was already quite evident[4]. Without my asking, Milton said his son was fat because his only food was breast milk. Milton's comment and Lucineide's story together show the high value placed on exclusive breastfeeding. On the other hand, Claudia's concerns show how difficult it is to maintain this practice.

Although health professionals are expected to make the encouraging of exclusive breastfeeding a  top priority, some observations at the health center indicate that they spend little time talking about the many aspects involved:

This exchange can be thought of as a moment of care where the doctor just gives her advice and barely listens to the woman who is seeking it. The (doctor's) project of the person seems to be restricted to just one person who is seen as needing to receive the appropriate information. The practical and emotional aspects are not contemplated.

Our discussion shows that, in line with recommendations from the Ministry of Health, the health professionals believe that breastfeeding is the best alternative for the child, without considering what the practical implications of this for women and their families may be. Some of our findings are similar to those of Azevedo et al.'s (2008) comparison of professionals' and mothers' points-of-view on breastfeeding, which noted that both groups consider that breastfeeding enhances immunity. Professionals and mothers differ, however, with respect to the causes of early weaning. According to the mothers, blame should be laid on the "weakness" of their milk and on the need to return to work. The professionals refer to lack of information - something that the team itself should offer - and to lack of interest.

In Prainha health professionals direct their interventions at mothers. Men are not included as a specific target for these interventions. They are expected to provide support, especially financial. However, among users, decisions relating to breast feeding are taken not only by the mother, but also by relatives, including other women, and also by partners - as in the case of Milton - although these persons are not included in health professionals' practices.

Users have separate and distinct concerns about the feasibility of exclusive breastfeeding.  Women and their families consider practical situations which they must deal with and which sometimes prevent them from breastfeeding their children. Yet they also feel it important to have access to health professionals and to be able to count on their care, so they do not openly declare "our side of the story". This shows that day-to-day healthcare practices involve negotiations that bring together and integrate the professional's point-of-view (providing guidance, following planned routines based on ideas that have support in biomedical knowledge) with the user's point-of-view (the need to solve practical problems and at the same time, to feel recognized by the health professionals). However, it is important to remember that care is constructed within unequal relations where professionals enjoy a superior position and dispose of major economic and social capital.

Final Thoughts

In this paper we have argued that care exists in all contacts between professionals and users, to the extent that within these exchanges projects of the person are constructed. As a result, the characteristics of care shift constantly. Nevertheless, we have also shown that the situations in which these encounters take place are structured in such a way that certain care practices are favored. Health professionals dispose of greater symbolic and economic capital and therefore their advice and practices are more highly valued than those of users.

We have shown that professionals prioritize the discursive aspects of care and seek to follow planned routines, displaying a tendency to devalue the behavior of users who do not comply with expectations. Instead of trying to understand the meaning of their behavior, professionals construe the 'do not follow the guidelines' as lack of interest or as an indicator of absence of family structure. Practices are built up from this perspective - for which we have coined the expression ‘partial project of the person' - where the person is assessed by considering just a single aspect. Desire - for example, in relation to pregnancy - or autonomy to decide about infant feeding are simply ignored. Thus, professionals offer quick consultations during which they proffer guidance without hearing the user's "side" of. the story.

When users engage with professionals they adopt their own strategies, giving ample space to spontaneity and physicality, an approach that contrasts with all the planning and the emphasis on discourse on the part of the professionals. Using these strategies users seek to give appropriate answers to practical problems. For this purpose they also consider the health professional's knowledge to be important. So, despite their differences, users do rate having access to the professionals highly and do recognize their high social value.

Thinking of care from the perspective proposed here, we could ask to what extent the lack of technical success - unplanned pregnancy or non-adherence to breastfeeding - implies the existence of problems. The way users care for themselves (turning up pregnant, not breastfeeding) can respond to practical needs or be part of strategies to deal with their own desires and with the limitations imposed on them by reality. To take this into account is central to the construction of health practices that might at once enable the building of genuine projects of happiness and of the person. At this point it should be evident that the concept proposed here and the work of Ayres stand in a complementary relationship to each other.

The development of care practices that adequately consider users' points-of-view demands including within interventions in the area of health, not just women's perspectives, but also those of members of the extended family. This includes not only spouses and other women (friends and kin), but also other male relatives such as uncles or brothers. This requires awareness that kinship cannot be conceived as restricted merely within consanguineal or co-residential relationships. Our work shows that when it comes to the participation of relatives in decisions relating to health care families differ among themselves. In some situations just the couple is involved (as in Paula and Ed's decision to abort), while in others co-resident relatives participate (such as in the decision to feed mingau to Thaddeus), and in still others a broad group of relatives and neighbors might take part (such the baby shower organized for Alicia).

This said, it is necessary to add that care cannot be thought of only from the perspective of exchanges between professionals and users. The concept of care proposed here includes socio-historical and cultural aspects. Thus, we must remember that health professionals and users both develop their practices within the context of very poor conditions, not just the working conditions of the professionals, but also in the face of insufficient infrastructure and a lack of working materials. This context makes it difficult to develop care practices that contemplate users' points-of-view and contributes to the broader frame - the high rates of maternal death and infant mortality that constitute Brazilian reality (Rattner, 2009).

REFERENCES

AQUINO, E.M. et al. Adolescência e reprodução no Brasil: a heterogeneidade dos perfis sociais.  Cad. Saúde Pública, v. 19, Sup 2, p. 377-388, 2003.

AYRES, J.R.C.M. Organização das Ações de Atenção à Saúde: modelos e práticas. Saúde Soc, v.18, Sup 2, p. 12-23,  2009.

AYRES, J.R.C.M. Sujeito, intersubjetividade e práticas de saúde. Ciênc. Saúde Coletiva, v.6, n.1 , p. 63-72, 2001.

AYRES, J.R.C.M. O Cuidado, os modos de ser (do) humano e as práticas de saúde. Saúde Soc. v.13, n.13, p.16-29, 2004b.

AZEREDO, C.M. et. al. Percepção de mães e profissionais sobre aleitamento materno: encontros e desencontros. Revista Paulistana de Pediatria, São Paulo, v.26, n. 4, p, 336-344,  2008.

BECKER, H. Métodos de pesquisa em Ciências Sociais. São Paulo: HUCITEC, 1994.

BEMFAM (Sociedade Civil Bem-estar Familiar no Brasil)/DHS (Demography and  Health Survey). Brasil Pesquisa Nacional sobre Demografia e Saúde. Rio de Janeiro: BEMFAM/DHS, 1997.

BOURDIEU, P. Razões práticas: sobre a teoria da ação. Campinas: Papirus, 1996.

BOURDIEU, P. O poder simbólico. Lisboa: Difel, 1989.

BRASIL. Ministério da Saúde. Agenda de compromissos para a saúde integral da criança e redução da mortalidade infantil. Brasília: Ministério da Saúde, 1997a.

BRASIL. Ministério da Saúde. Guia prático do Programa de Saúde da Família. Brasília: Ministério da Saúde, 2002.

BRASIL. Ministério da Saúde. Saúde da família: uma estratégia para a reorganização do modelo assistencial. Brasília: Ministério da Saúde, 1997b.

BUSTAMANTE, V. Cuidado infantil e construção social da pessoa: uma etnografia em um bairro popular de Salvador. 2009. PhD Thesis - Postgraduate Program in Collective Health, Federal University of Bahia.

CARVALHO, M.F. et. al. Acompanhamento do crescimento em crianças menores de um ano: situação nos serviços de saúde em Pernambuco, Brasil. Cad. Saúde Pública,  v. 24, n.3, p. 675-85, 2008.

CARVALHO, M. P de. No coração da sala de aula: gênero e trabalho nas séries iniciais. São Paulo: Xamã, 1999.

DALSGAARD, A.L. Vida e esperanças: esterilização feminina no Nordeste. São Paulo: Editora UNESP, 2006.

FROTA, M.A.; BARROSO, M.G.T. Repercussão da desnutrição infantil na família. Rev. Latino-am enfermagem, v.13, n.6, p. 996-1000, 2007.

GEERTZ, C. El antropólogo como autor. Buenos Aires: Paidós, 1989.

JACKSON, M. (Ed.) Things as they are. New directions in phenomenological anthopology. Bloomington: Indiana University Press, 1996.

MENEZES, G.; AQUINO, E.M.  Pesquisa sobre aborto no Brasil: Avancos e desafios para o campo de genero, sexualidade e saúde reprodutiva. In: Seminário Estudos sobre a Questão do Aborto em Países da América do Sul, com ênfase no Brasil. Campinas, 2007.

MENEZES, G. Aborto, juventude e saúde. 2006. PhD Thesis Postgraduate Program in Collective Health, Federal University of Bahia.

MOURA, M.L.S. de et. al. Conhecimento sobre desenvolvimento infantil em mães primíparas de diferentes centros urbanos do Brasil. Estud. Psicol. (Natal), v.9,n.3, p. 421-29, 2004.

OLIVEIRA, M.I.C. de; CAMACHO, L.A.B. Impacto das Unidades Básicas de Saúde na duração do aleitamento materno exclusivo. Rev. bras. Epidemiol, v.5, n.1 , p. 41-51, 2002.

RABELO, M.C. A experiência de indivíduos com problema mental: entendendo projetos e sua realização. In: RABELO, M.C.; ALVES P.C.; SOUZA, I.A. (Orgs.). Experiência de doença e narrativa. Rio de Janeiro: Editora Fiocruz, 1999. p.205-227.

RABUSKE, M.M.; OLIVEIRA, D.S.; ARPINI, D.M. A criança e o desenvolvimento infantil na perspectiva de mães usuárias do serviço público de saúde. Estud. Psicol. (Natal), v.22, n.3, p. 321-31, 2005.

RATIS, C. de A.S.; BATISTA FILHO, M. Aspectos estruturais e processuais da vigilância do crescimento de menores de cinco anos em serviços públicos de saúde do estado de Pernambuco. Rev. Bras. Epidemiol, v.7, n.1, p. 44-53, 2004.

RATTNER, D. Humanização na atenção a nascimentos e partos: breve referencial teórico. Interface - Comunic, Saúde, Educ, v.13, supl.1, p.595-602, 2009.

SANTOS, S.R.. et. al. Avaliação da assistência à saúde da mulher e da criança em localidade urbana da região sudeste do Brasil. Rev. Saúde Pública, v. 34, n.3, p. 266-71, 2000.

SCAVONE, L. O trabalho das mulheres pela saúde: cuidar, curar, agir. In: VILLELA, W; MONTEIRO, S. (Orgs.). Gênero e Saúde Programa Saúde da Família em questão. Rio de Janeiro: Abrasco, 2005.

SCHRAIBER, L.B. Equidade de gênero e saúde: o cotidiano das práticas no Programa de Saúde da Família. In: VILLELA, W.; MONTEIRO, S. (Orgs.). Gênero e saúde: Programa de saúde da família em questão. Rio de Janeiro: Abrasco, 2005. p.39-61.

SIQUEIRA, M. J. et. al. Profissionais e usuárias (os) adolescentes de quatro programas públicos de atendimento pré-natal da região da grande Florianópolis: onde está o pai?. Estud. Psicol. (Natal), v.7, n.1, p. 65-72, 2002.

SLOMP, F.M. et. al. Assistência ao recém nascido em um Programa de Saúde da Família. Rev. esc. enferm. USP, v.41, n.3, p. 441-46, 2007.

THOMAS, C. De-constructing concepts of care. Sociology, Inglaterra,  v.27, n.4, p.649-669,  nov, 1993.

TOREN, C. (1997). Ethnography: theoretical background. In: J.T.E Richardson (Ed). Handbook of Qualitative Research Methods for Psychology and the Social Sciences. London: BPS.

VICTORA, C. et. al. Estudo longitudinal da população materno-infantil da região urbana do Sul do Brasil, 1993: aspectos metodológicos e resultados preliminares. Rev. Saúde Pública, v.30, n.1, p. 34-45, 1997.

i Address: Institute of Collective Health (ISC), UFBA, Rua Basílio da Gama, s/n, Salvador, BA, Brazil. 40.110-170.
1 All names used here are fictitious. ]]> 2 In colloquial Portuguese the expression aparecer grávida means both ‘to be seen to be pregnant' and ‘to turn up pregnant'.
3 The fieldwork was conducted by Vania Bustamante alone. For this reason, we occasionally use the first person with reference to some of our material.
4 Vania Bustamante continued fieldwork throughout her pregnancy.