Lilia Blima SchraiberI; Ana Flávia P L D'OliveiraI; Ivan França-JuniorII; Simone DinizIII; Ana Paula PortellaIV; Ana Bernarda LudermirV; Otávio ValençaVI; Márcia Thereza CoutoI
IDepartamento de Medicina Preventiva. Faculdade de Medicina. Universidade de São Paulo (USP). São Paulo, SP, Brasil
IIDepartamento de Saúde Materno-Infantil. Faculdade de Saúde Pública. USP. São Paulo, SP, Brasil
IIIColetivo Feminista, Sexualidade e Saúde. São Paulo, SP, Brasil
IVInstituto Feminista para a Democracia. Recife, PE, Brasil
VDepartamento de Medicina Social. Centro de Ciências da Saúde. Universidade Federal de Pernambuco. Recife, PE, Brasil
VICoordenadoria de Pesquisa e Extensão. Faculdade de Ciências Médicas de Pernambuco. Recife, PE, Brasil
OBJECTIVE: To analyze the results from the "WHO Multi-country Study on Women's Health and Domestic Violence", on the prevalence of intimate partner violence against women found in Brazil.
METHODS: This cross-sectional study was part of the "WHO Multi-country Study on Women's Health and Domestic Violence against women", which was carried out in ten countries between 2000 and 2003. All the countries used a standardized structured questionnaire, devised for this study. In order to obtain data from contrasting settings within each country, the biggest city and a rural region were investigated whenever feasible. A representative sample of women aged 15 to 49 years was selected from the city of São Paulo and 15 municipalities in a rural region of the northeast, the Zona da Mata de Pernambuco. The study included 940 women from São Paulo and 1,188 from Zona da Mata de Pernambuco who had had an intimate partner at some time in their lives. Violence was classified as psychological, physical and sexual types, and was analyzed in relation to overlapping, recurrence of episodes, severity and when it occurred.
RESULTS: The women in São Paulo and Pernambuco respectively reported the following at least once in their lifetimes: psychological (N=383; 41.8% and N=580; 48.9%); physical (N=266; 27.2% and N=401; 33.7%); sexual (N=95; 10.1% and N=170; 14.3%) violence. There was significant overlapping among the types of violence, which seemed to be associated with the most severe types of violence. The greatest single type was psychological violence, in São Paulo and Pernambuco (N=164; 17.5% and N=206; 17.3%), and the smallest was sexual violence (N=2; 0.2% and N=12; 1.0%).
CONCLUSIONS: The results show that violence is a very common phenomenon. The findings reiterate previous international studies results with regard to high magnitude and overlapping of types of intimate partner violence.
Key words: Battered women. Violence against women. Spouse abuse. Domestic violence. Cross-sectional studies.
Violence experience by women is a health problem nowadays. The commonest form is the violence perpetrated by intimate partners. International organizations (World Health Organization WHO, Pan-American Health Organization PAHO and World Bank)9,14 and a health professionals' organization (American Medical Association)1 have published data showing high prevalences of intimate partner violence. As well as this high magnitude, these episodes of violence are often of a severe and repeated nature,2,7,17,22 thereby expressing inequality in gender relations. The repercussions extend to women's physical, psychological and reproductive health and may remain even after the violence ceases.18,19
The existing population-based data are difficult to compare, since the studies were carried out on different populations, using different instruments, conditions of privacy, interviewer training and information collection techniques, and also with different definitions of violence. Most studies have only focused on physical violence, generally measured in terms of concrete actions such as slaps, punches and shoves. In 48 population-based surveys carried out around the world, between 10% and 69% of the women reported that they had suffered at least one episode of physical violence from a partner during their lives.8 The percentage of the women who had been attacked during the year preceding the interviews ranged from 3% to 27%.8
Sexual and psychological violence perpetrated by an intimate partner has been little investigated and the information is even more imprecise. Several factors contribute towards making it difficult to recognize and delimit sexual violence within stable partner relations. For example, different names are given to acts of aggression (violence, rape, abuse and, sometimes, harassment), and also in many cultures non-consensual sexual acts are considered to be the wife's duty.
Likewise, the definition of psychological violence varies widely between women and men in different cultures, thereby giving rise to difficulties in defining and consequently measuring it. However, the surveys have indicated that physical violence is usually accompanied by psychological violence and, in one third to half of the cases, also by sexual violence.2,7,8,13
In Brazil, a population-based study measuring the occurrence of violence against women21 was carried out on a representative national sample of 2,502 women aged 15 years and over. In this investigation, 43% of these Brazilian women declared that they had suffered violence inflicted by a man at some time in their lives; one third said they had suffered some form of physical violence, 13% sexual violence and 27% psychological violence. Husbands, former husbands, boyfriends and former boyfriends were the main aggressors, accounting for between 88% of the slaps and shoves to 79% of the acts of forced sexual intercourse.
With the aim of filling the gap in the population-based data on the different forms of intimate partner violence, the WHO Multi-country Study on Women's Health and Domestic Violence4-6 was proposed by the World Health Organization and was carried out in ten countries (Brazil, Peru, Bangladesh, Japan, Thailand, Samoa, Namibia, Ethiopia, Serbia and Tanzania). Making use of the same methodology and data collection instruments, home surveys were conducted in one metropolis and in rural regions or provinces in five of the countries, while surveys were only conducted in large cities in the other five countries. The prevalence and factors associated with different forms of violence against women aged 15 to 49 years were analyzed, along with the impact of this violence on the health of women and children and the ways of dealing with it that were adopted.
The present article analyzes the results from this multi-country study in relation to the prevalence of intimate partner violence found in Brazil.
A cross-sectional study was planned so as to construct representative samples of women aged 15 to 49 years who lived in the city of São Paulo (SP) and in the rural region of Zona da Mata, in the State of Pernambuco (ZMP), with the exclusion of the metropolitan region of Recife. The data collection took place between 2000 and 2001. To standardize the design and with the aim of obtaining data from different situations within each country, whenever possible, the largest city and another region with more rural characteristics were surveyed. As an ethical requirement of the study protocol, both the locations chosen needed to have services capable of receiving cases of women who were the victims of violence identified during the investigation.
The strategy of multiple-stage cluster sampling was used.12 The main sampling procedures are described in another article.20 In São Paulo, a probabilistic matrix drawn up by the Instituto Brasileiro de Geografia e Estatística (IBGE Brazilian Institute for Geography and Statistics) for the Pesquisa Nacional por Amostragem de Domicílios (PNAD National Household Sampling Survey) in 1995 was used. This was composed of 263 census tracts that were selected with probability proportional to size (PPS). Selection of the sampling units was done in three stages, with PPS: the first consisted of drawing 72 census tracts from the 263 tracts in the matrix, ranked according to the proportion of heads of household with less than one year of schooling, and a systematic draw was made from this relationship. Out of the 72 tracts drawn, the outlines of 40 of them were updated using the PNAD 1999 register.* In the second stage, a fixed number of 30 homes were drawn systematically in each census tract. Lastly, for each home where women aged 15 to 49 years were living, only one woman was drawn for interviewing.
In the ZMP, the selection of the sampling units took place in four stages: first, the 42 municipalities in the ZMP were ranked according to demographic density, urbanization rate and illiteracy rate among the heads of household, thus resulting in systematic selection of 15 municipalities with PPS. In the second stage, eight census tracts were selected from each municipality, thus totaling 120 tracts. In the third stage, a sample of 18 homes was selected from each census tract, and lastly, only one woman aged 15 to 49 years was drawn from each household.
These draws resulted in 2,163 homes in SP and 2,136 in the ZMP. Among these addresses, 347 (16%) in SP and 180 (8.4%) in the ZMP were not located. Also, 412 in SP (24%) and 401 (20.7%) in the ZMP did not have any women aged 15 to 49 years. Thus, 1,303 eligible women were found in SP and 1,539 in the ZMP, giving 1,172 completed interviews in SP (90%) and 1,473 (95.7%) in the ZMP. The refusal rates among the eligible women who were approached were low: 3.7% in SP and 1.2% in the ZMP.
Intimate partners were defined as the male companion or former companion, independent of whether there was any formal union, and present boyfriends provided that there was a sexual relationship. Among the women who completed the questionnaire, 940 women in SP and 1,188 in the ZMP fulfilled the condition of having had an intimate partner during their lifetimes.
The home interviews used a standardized questionnaire that had been constructed by the international team with collaboration from the investigators in the participating countries. The questionnaire was formulated in English, translated into Portuguese and back translated into English, with verification by the WHO team itself.6 Before arriving at the final version of this questionnaire, the questions on violence were adapted to the national culture with the aid of qualitative surveys (16 focus groups; 12 in-depth interviews and 39 interviews with key informants) that were carried out among men, women and service professionals both in SP and in the ZMP. The questionnaire was discussed by consultative committees that were set up to follow the survey process, and it was pretested with regard to its clarity and ease of application. In analyzing its internal consistency, the Cronbach's values for psychological, physical and sexual violence were respectively 0.784, 0.827 and 0.778 for SP and 0.793, 0.829 and 0.772 for the ZMP. In the pilot study carried out in both locations, the instrument was shown to be capable of discriminating between the two sociocultural contexts.
Intimate partner violence (psychological, physical or sexual) was considered to be present when the woman answered yes to at least one of the items in the respective block of questions. The physical violence was considered to be moderate when only items 1 and 2 were affirmed and severe for the other items, in accordance with the definition in the protocol for the international study.6 Each episode was explored with regard to whether it occurred before or during the 12 months preceding the interview date, and whether it was repeated. Recurrence of the episodes was noted according to the interviewee's perception, in terms of one, a few or many times.
The teams of interviewers (who were all women) were specifically trained for the survey, for one month.
Special attention was paid to ethical issues, including a concern for the personal safety of the interviewees and interviewers, because of the complex and sensitive nature of this topic. Confidentiality and security were preserved through several measures: a guarantee of privacy during the home interview (only children of less than two years old were allowed into the interview location); the use of transportation vehicles adapted for carrying out the interview, in the event that the home had no privacy or was unsafe for the woman; no personal identification of the interviewee in the questionnaires; an alternative questionnaire to be used in the event of a threat to privacy (arrival of the partner or other people at the interview location); strategies for greater security in the logistics of access to homes, such as working in teams in which the members were geographically close to each other and with the use of mobile phones for rapid communication, including with the central office for the survey; and psychological support for the interviewers and supporting care for the interviewees. As well as taking part in the training and supervision of the teams, the psychological support for the interviewers and the first port-of-call for the interviewees in case of need included another figure: counselors. These were selected from among people with experience of dealing with the problem of domestic violence against women. In hiring drivers, the individuals selected were those who, after receiving brief guidance, showed themselves to be capable and sensitive with regard to supporting the teams, in addition to being able to drive safely and having knowledge of the places visited.
The support care was developed through partnerships with specialized services caring for women who are victims of violence, in which it was agreed in advance that they would receive cases that might be found during the survey. To identify such services, a detailed survey of the existing care networks for victims of violence in São Paulo and Pernambuco was carried out. From this, small-format service guides (mini-guides) were drawn up, indicating the different types of assistance provided (police, legal, psychological or medical), for distribution among the interviewees as educative material and support for the care needs in future situations. These mini-guides, along with calendars containing information on human rights and women's rights and indications of dates that commemorate these rights, were distributed to the interviewees independent of whether they reported any violence, by way of educative material for the survey.
At the beginning of the interview, each participant was informed about the study and was told that the interview could be halted at any time, and her consent was requested.
For interview quality control, new interviews were conducted on 10% of the sample using a summarized questionnaire, to check for consistency. Furthermore, there was weekly supervision and monitoring of the response rate obtained by each interviewer, and comparison between them. To reduce the losses in São Paulo, presentation letters were sent to the homes that had been drawn, prior to the visits by the teams. If nobody was found in the homes selected and/or the selected women were not found on the day of the first visit, another two return visits were made.
Since the sample elements had unequal selection probabilities (census tracts with different growth rates from 1995 to 2000 and different numbers of eligible women in each home), there was a need for weighted analysis. However, since the weighted data did not differ significantly from the raw data, the results were presented without weighting.
The data are presented in the form of means and proportions, using Student's t statistics and Pearson's chi-squared to test for possible differences in the means and proportions, respectively. Differences with <5% were deemed to be statistically significant.
The study was approved by the Research Ethics Committee of the Faculdade de Medicina of Universidade de São Paulo (CAPPesq-609/98) on 11/11/1998, and by the National Research Ethics Commission (Report No. 002/99) on 1/ 11/ 1999.
Some sociodemographic characteristics of the interviewees are presented in Table 1. The women living in SP had higher schooling levels, smaller numbers of children, more often had their own income and more often were married or had boyfriends than did the women in the ZMP.
At least one form of intimate partner violence was reported by 436 women in SP (46.4%; 95% CI: 43.2%;49.6%) and 644 women in the ZMP (54.2%; 95% CI: 51.4%;57.1%). These prevalences are presented in Table 2 according to type of violence and time when it occurred. The overlaps between types of violence are shown in Figure 1 (lifetime) and Figure 2 (year preceding interview).
Psychological violence was the most frequent event over the women's lifetime and also during the year preceding the interview. The episodes were not single occurrences: only 90 (22.9%) in SP and 104 (17.9%) in the ZMP reported that psychological violence had occurred only once in their lifetime; 157 (40%) in SP and 255 (44%) in the ZMP reported many episodes and the remainder said that such episodes occurred a few times.
In 90% of the cases, psychological violence accompanied the reports of physical or sexual forms of violence. In fact, among the women who reported some form of physical or sexual violence, only 43 women in SP (10%) and 64 in the ZMP (9.9%) did not also report some episode of psychological violence.
However, in contrast with exclusively physical or sexual violence, psychological violence alone occurred at a much higher rate: 37.6% of the cases in SP and 32% in the ZMP of episodes during the women's lifetime and approximately half in the year preceding the interview (Figures 1 and 2).
When occurring alone, psychological violence seemed to be expressed more moderately than when associated with other forms of violence. Among the women who declared that that had solely suffered psychological violence, the episodes most reported were insults alone (41.5% in SP and 30.7% in the ZMP). Only five women in SP (3.05%) and three in the ZMP (1.46%) who reported exclusively psychological violence said that they had suffered all its expressions (insults, humiliation, intimidation and threats). On the other hand, among the women who said that they had suffered all three forms of violence (physical, psychological and sexual), 34 in SP (45.33%) and 70 in the ZMP (53.85%) reported suffering from all four expressions of psychological violence, while only two in SP (2.97%) and three in the ZMP (1.46%) reported only insults.
Slaps and shoves were the most frequent acts reported with regard to physical violence (Table 2). For 12.4% of the interviewees in the ZMP, an intimate partner had threatened them with or actually used a weapon on them at some time during their lives.
Regarding recurrence of the episodes, physical violence had occurred once for 37.9% of the women in SP and 35.2% in the ZMP. Multiple episodes were reported by 23.4% of the women in SP and 32.7% in the ZMP. The remainder declared that such episodes occurred a few times.
The majority of the episodes were considered to be of greater severity (items 3 to 6 of physical violence; Table 2): 57% (146/256) of the cases in SP and 59.3% (237/401) in the ZMP.
There were a greater number of reports of sexual and psychological violence among the women who declared they had suffered severe physical violence than among those declaring moderate physical violence. Among the women who reported severe physical violence, 91.8% in SP and 92% in the ZMP also reported some form of psychological violence, while among those who reported moderate physical violence these figures were 73.6% in SP and 79.8% in the ZMP. Among those who had not suffered physical violence, the prevalence of psychological violence was 26.1% in SP and 29.4% in the ZMP. When severe physical violence occurred, simultaneous episodes of sexual violence affected 39.7% of the women in SP and 43.9% in the ZMP. In the cases in which moderate physical violence was reported, the occurrence of sexual violence was lower: 19.1% in SP and 17.8% in the ZMP. In the cases in which no physical violence was reported, 2.3% of the women in SP and 4.3% in the ZMP declared that they had suffered sexual violence.
Sexual violence occurred least frequently, but almost always together with other types of violence (Figures 1 and 2). Among the women who reported occurrences of sexual violence, 82.1% in SP and 71.8% in the ZMP reported that their partners used physical force to have sexual intercourse. Also among the women who declared they had suffered sexual violence, 69.5% in SP and 67.6% in the ZMP had sexual intercourse for fear of what their partner might do. Degrading or humiliating sexual practices were the types least cited. With regard to recurrence, 26 women in SP (27.4%) and 37 in the ZMP (21.8%) reported one episode; 31 in SP (32.6%) and 67 in the ZMP (39.4%) reported many episodes; and the remainder declared that there had been a few episodes.
The present study is the first specially designed Brazilian population-based study on violence against women that enables comparison with other countries.
One potential limitation of this study is the low precision of the information relating to intimate partner violence. With backing from the literature,3 it can be accepted that, in the context of an interview, it is unlikely that women would report episodes of violence that had not occurred, because of the stigma attached to such episodes and their relationship with shame and blame. On the other hand, it is possible to suppose that events of this nature may often be concealed, for the same reasons. In the present research, unexplored data regarding underreporting of sexual abuse during infancy was found in the face-to-face interviews. This underreporting could be of the order of 40%, in relation to forms filled out anonymously.16 The training and careful supervision for the fieldwork, the translation and application of the pilot questionnaire and the detailed series of ethical precautions sought to minimize as far as possible the difficulties in revealing such episodes. There may nonetheless have been underreporting in the information collected and thus underestimation of the prevalence.
Even though the questionnaire in Portuguese was preceded by qualitative studies for cultural adaptation, it may have generated some information bias, especially in the ZMP, because of the lower level of schooling there.
The sampling design and the small number of refusals indicate that these samples were representative, thus minimizing the potential selection bias. Comparison of the age group of the sample obtained with the age group of the population in the study locations contributed towards assuring the trustworthiness of the data relating to the population group investigated.5
The rates of intimate partner violence found for Brazil are not among the highest rates when compared with data from other countries.2,7,8,9,11,22 Among the 15 localities studied with the same methodology in the multicenter study,5 ranked according to the prevalence found, there was a range of physical violence from 13% (Japan) to 61% (Peru), and a range of sexual violence from 6% (Japan and Serbia/Montenegro) to 59% (Ethiopia). Thus, Brazil was in an intermediate or low position within this range, although the ZMP always presented higher rates than seen in SP.6
Rural localities generally presented higher rates than did urban localities.5 Several explanations may be put forward for this fact, for example gender relations in urban regions that are more distant from traditional patterns which has been suggested by studies on the factors associated with intimate partner violence8,10,11 and greater presence of women's movements and support services in urban regions. Subsequent studies may provide greater depth of explanation for the differences encountered.
In the case of Brazil, almost three decades of activism in confronting violence against women and in developing institutional responses may have contributed towards making such violence more visible and less acceptable. Nonetheless, this does not mean that acts of violence have ceased, as indicated by the high rates encountered.
However, activism may be the reference point for explaining the relatively high levels of psychological violence in Brazil, in relation to other forms of violence and to other countries.6 In other words, the lower acceptability of intimate partner violence might in general lead to greater sensitivity and resultant revelation of psychological violence, in relation to other countries. Cultural differences regarding ease of expression may also be responsible for this finding.
The prevalences found in the present study are even greater than in another study carried out in Brazil.21 A recent study15 has also revealed high prevalences in this country, although those measurements referred to estimates of the last 12 months and to violence between the couple (either partner as the aggressor).
High occurrence of cases was observed, including recurrent and severe cases. In the year 2000, the Brazilian census found that there were 3,135,015 women aged 15 to 49 years living in the city of São Paulo.** Since the data for the present study were collected at the end of 2000 and beginning of 2001, it may be inferred by extrapolating the percentages to the general population that, in 2000, 586,248 women in the municipality of SP may have suffered some episode of psychological violence, 260,206 some form of physical violence and 87,780 sexual violence, committed by an intimate partner. For the same period, there were 316,881 women aged 15 to 49 years in the ZMP, and thus it can be estimated that in this locality 153,953 women suffered psychological violence, 106,789 physical violence and 45,314 sexual violence.
The findings from the present study are also consistent with those in the literature in indicating that most of the physical violence is accompanied by psychological violence and that, in turn, sexual violence is usually associated with physical violence.2,13,17 Attention is drawn to the small number of cases in which sexual violence was not accompanied by physical aggression, for example the cases in which the woman felt coerced by "fear of what he might do". One possible explanation for this lies in the culture of a dominant gender and the low perception of such duress within relationships of conjugal type. Sexual violence alone was a rare finding in most of the countries researched, although with some exceptions, such as the 8.4% of the women in Haiti who reported sexual violence alone** and considerable proportions in the capital of Thailand and rural regions of Ethiopia and Bangladesh, where this form corresponded to 30% of the cases in which there was some physical or sexual violence.5
The present study has revealed that intimate partner violence may have different expressions: in the two Brazilian regions studied, the most frequent situation was the occurrence of psychological violence alone, followed by physical violence accompanied by sexual violence and then by the three forms together. The most severe violence seemed to be associated with greater overlapping of the different forms of violence: associated psychological, physical and sexual violence accounted for approximately 20% of the cases, and this constitutes an extremely serious situation.
As discussed in the most recent literature,8 these data suggest that the phenomenon of violence may present with at least two defined patterns: more moderate situations in which various conflicts or episodes of frustration and rage occasionally erupt as aggression; and a more serious and chronic pattern of progressive nature. The three most frequent types of expression (psychological violence alone; psychological and physical violence; and the three forms together) need to be the target of future studies, so that possible differences between them can be investigated. Such differences might be explained in terms of associated factors and repercussions for health, as well as their behavior over time in terms of any evolution of more moderate forms to more severe forms.
Such information will be useful for interventions, since it is generally the most severe cases that seek help from specialized services. On the other hand, very few acute cases or cases of more occasional nature are picked up by primary healthcare units and schools, for example. Detection at these places might be an important means of preventing or reducing the more severe forms.
The findings regarding the time when the violence took place are of concern: the lifetime prevalence was more than twice what was found for the year preceding the interview. Present violence may indicate situations that have recently emerged, or well-established and chronic situations. The greater number of episodes prior to the last year before the interview may indicate the possibility that there is a way out from the situation of violence. In Ethiopia, 54% of the women reported that they had suffered physical and/or sexual violence committed by their partners over the preceding year, versus 17% who reported episodes prior to the last year, thus showing a situation that is the inverse of what was seen in Brazil.5 These data may indicate that the reality in Ethiopia, which has higher prevalence of violence, is one of greater continuation of situations of violence. New studies might be able to go into this question in greater depth, through seeking the factors associated with remaining in or going out from situations of violence.
The results presented describe a phenomenon that is very common but still little known: violence against women committed by their intimate partners, as revealed by its great magnitudes of prevalence, severity and recurrence. Furthermore, even if new analyses and studies are recommended, the knowledge now available makes it legitimate to consider that there is an urgent need to implement, supervise and evaluate public policies that are specially directed towards intimate partner violence.
To Ricardo Góes of the Department of Preventive Medicine (DMPR) of the Faculdade de Medicina of Universidade de São Paulo (FMUSP), for logistic support for the survey; to Andréia Nascimento (DMPR/FMUSP) for assistance in the data analysis.
1. American Medical Association. Diagnostic and guidelines on domestic violence. Chicago; 1992.
2. Ellsberg M, Peña R, Herrera A, Liljestrand J, Winkvist A. Candies in hell: women´s experience of violence in Nicaragua. Soc Sci Med. 2000;51(11):1595-610.
3. Ellsberg MC, Heise L, Peña R, Agurto S, Winkvist A. Researching domestic violence against women: methodological and ethical considerations. Stud Fam Plann. 2001; 32(1):1-16.
4. Garcia-Moreno C, Watts C, Jansen H, Ellsberg M, Heise L. Responding to violence against women: WHO´s Multi-country: Study on Women's Health and Domestic violence. Health Hum Rights. 2003;6(2):113-27.
5. Garcia-Moreno C, Jansen HA, Ellsberg M, Watts CH. WHO multi-country study on women´s health and domestic violence against women: initial results on prevalence, health outcomes and women´s response. Geneva: World Health Organization; 2005.
6. Garcia-Moreno C, Jansen HA, Ellsberg M, Heise L, Watts CH et al. Prevalence of intimate partner violence: findings from the WHO multi-country study on women's health and domestic violence. Lancet. 2006;368(9543):1260-9.
7. Heise L, Ellsberg M, Gottemoeller M. Ending violence against women. Baltimore: Johns Hopkins University School of Public Health, Population Information Program; 1999. (Population reports, Series L, 11).
8. Heise L, Garcia-Moreno C. Intimate partner violence. In: Krug EG, Dahlberg LL, Mercy JA, Zwi AB, Lozano R. (editors). World report on violence and health. Geneva: World Health Organization; 2002. p.91-121.
9. Heise L, Pitanguy J, Germain A. Violence against women: the hidden health burden. Washington: The International Bank for Reconstruction and Developement/The World Bank; 1994.
10. Jewkes R, Levin J, Penn-Kekana L. Risk factors for domestic violence: findings from a South African cross-sectional study. Soc Sci Med. 2002;55(9):1603-17.
11. Jewkes R. Intimate partner violence: causes and prevention. Lancet. 2002;359(9315):1423-9.
12. Kish L. Survey sampling. New York: John Wiley & Sons; 1965.
13. Kronbauer JFD, Meneghel SN. Perfil da violência de gênero perpetrada por companheiro. Rev Saude Publica. 2005;39(5):695-701.
14. Organizacion Panamericana de la Salud. Violencia contra la mujer: un tema de salud prioritario. Washington (DC): Division de Salud Familiar y Reproductiva; Division de Salud y Desarrollo Humano; 1998.
15. Reichenheim ME, Moraes CL, Szklo A, Hasselmann MH, Souza ER, Lozana JA et al. The magnitude of intimate partner violence in Brazil: portraits from 15 capital cities ant the Federal District. Cad Saude Publica 2006;22(2):425-37.
16. Schraiber LB, Oliveira AFPL, França-JR I, Diniz CS, Portella AP, Ludermir AB, et al. Visibilidade/invisibilidade no estudo epidemiológico da violência sexual contra a mulher (VSCM) na cidade de São Paulo e Zona da Mata de Pernambuco. Rev Bras Epidemiol. 2002;(supl. esp.):36.
17. Schraiber LB, Oliveira AFPL, França-Junior I, Pinho A. Violência contra a mulher: estudo em uma unidade de atenção primária à saúde. Rev Saude Publica. 2002;36(4):470-7.
18. Schraiber LB, Oliveira AFPL. Violência contra mulheres: interfaces com a saúde. Interface - Comunic Saude Educ. 1999;3(5):11-27.
19. Schraiber LB, Oliveira AFPL, Falcão MTC, Figueredo WS. Violência dói e não é direito: a violência contra a mulher, a saúde e os direitos humanos. São Paulo: Editora da UNESP; 2005.
20. Silva NN, Cunha TN, Quintanilha JA. Amostra mestra e geoprocessamento: tecnologias para inquéritos domiciliares. Rev Saúde Pública. [periódico na Internet]. [Acesso em 15 jul 2007]; 37(4):494-502. Disponível em: http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0034 -89102003000400015&lng=pt&nrm=iso.
21. Venturi G, Recamán M, Oliveira S, organizadores. A mulher brasileira nos espaços público e privado. São Paulo: Fundação Perseu Abramo; 2004.
22. Watts C, Zimmerman C. Violence against women: global scope and magnitude. Lancet.2002;359(9313):1232-7.
Lilia Blima Schraiber
Grupo de Pesquisa e Intervenção
Violência e Gênero nas Práticas de Saúde
Departamento de Medicina Preventiva
Faculdade de Medicina Universidade de São Paulo
Av. Dr. Arnaldo 455 2ºandar - Cerqueira César
01246-903 São Paulo-SP
Pesquisa financiada pelo Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq Projeto Integrado; Proc. 523348/96-7) e Ministério da Saúde / Programa Nacional de DST/AIDS (Ref: 914 BRA 59 DST-AIDS II; ED 00/4772; Unesco 914/BRA/59).
Estudo produzido a partir dos dados da pesquisa "WHO MultiCountry Study on Women's Health and Domestic Violence against women", coordenado e financiado pela Organização Mundial de Saúde.
* Instituto Brasileiro de Geografia e Estatística. Pesquisa nacional de amostragem por domicílios; 1999. [Acesso em 25 ago 2001]. Disponível em: http://www.ibge.gov.br
** Kishor S, Johnson K. Profiling domestic violence: a multi-country study. Caverton (maryland): OCR Macro; 2004. [acesso em 15 jul 2007]. Disponível em http://www.measuredhs.com/pubs/pdf/OD31/OD31.pdf