Pergamon

Child Abuse & Neglect, Vol. 23, No. 11, pp. 11171126, 1999 Copyright 1999 Elsevier Science Ltd Printed in the USA. All rights reserved 0145-2134/99/$see front matter

PII S0145-2134(99)00078-2

PREVALENCE AND EFFECTS OF CHILD SEXUAL ABUSE IN A POOR, RURAL COMMUNITY IN EL SALVADOR: A RETROSPECTIVE STUDY OF WOMEN AFTER 12 YEARS OF CIVIL WAR
LINDA M. BARTHAUER
Department of Pediatrics, University of Rochester Medical Center, Rochester, NY, USA

JOHN M. LEVENTHAL
Department of Pediatrics, Yale University School of Medicine, New Haven, CT, USA

ABSTRACT
Objective: The purpose of this retrospective study was to examine the epidemiology of child sexual abuse (CSA) among women in a poor, rural community in El Salvador, which was recovering from a 12-year civil war. Methods: A cross-sectional, door-to-door survey was administered to 83 women. The LA Times Sexual Abuse Survey was used to determine the prevalence of CSA. The Hopkins Symptom Checklist (HCL) was used to measure long-term psychological sequelae of abuse (somatization, anxiety, depression, interpersonal sensitivity, and obsessive-compulsive traits). Information also was obtained about participation and number of relatives killed in the countrys 12-year civil war. Results: Of the 83 women interviewed, 14 (17%) reported a total of 21 experiences of CSA. The median age of abuse was 14 years. The majority of perpetrators (90.4%) were strangers, friends, or neighbors. None was a parent. The only signicant difference between abused and non-abused women was on the depression measure, where abused women showed more pathology. After controlling for the number of relatives killed in the war, however, that difference failed to reach statistical signicance. Conclusions: Differences in the epidemiology of CSA from that of other countries may be secondary to a different social structure in the rural Salvadoran community and non-disclosure by the women surveyed. The lack of difference in psychological symptoms between abused and non-abused women may be related to the different characteristics of the abuse and perpetrators. It also may be secondary to adverse social conditions such as poverty and war, which could obscure the long-term effects of CSA. 1999 Elsevier Science Ltd Key WordsEl Salvador, Child sexual abuse, Epidemiology, Symptoms.

INTRODUCTION ALTHOUGH THE SEXUAL abuse of children has been recognized as a major problem in developed countries over the last two decades, there have been few systematic studies of child sexual abuse (CSA) in developing countries. There is little question that CSA is an international

This research was support in part by a Resident Research Grant from the American Academy of Pediatrics and by the Department of Pediatrics, Yale University School of Medicine. Submitted March 10, 1997; nal revision received November 11, 1998; accepted November 21, 1998. Requests for reprints should be sent to Linda M. Barthauer, MD, MPA, Department of Pediatrics, University of Rochester Medical Center, PO Box 777, 601 Elmwood Avenue, Rochester, NY 14642. 1117

1118

L. M. Barthauer and L. M. Leventhal

phenomenon (Finkelhor, 1994; Finkelhor & Korbin, 1988). In a 1994 review of the international epidemiology of CSA by Finkelhor, retrospective surveys of women in 21 countries, 19 of them industrialized, yielded prevalence rates of 7 to 36%. Some clear patterns also emerged from this review. In all countries, the offenders were overwhelmingly male, and intrafamilial abuse accounted for approximately one third of the episodes of abuse (range 14 to 56%). Studies published since the review have conrmed these patterns (Bendixen, Muus, & Schei, 1994; Goldman & Padayachi, 1997; Halperin et al., 1996; MacMillan et al., 1997; Tonmyr, 1998). The long-term psychological effects of CSA have been examined in industrialized countries, mostly in studies originating in North America. Symptoms associated with a past history of CSA have included substance abuse, sexual problems, somatization, anxiety, depression, dissociation, suicidality, and poor self-esteem (Bendixen et al., 1994; Briere & Runtz, 1988; Browne & Finkelhor, 1986; Cheasty, Clare, & Collins, 1998; Jumper, 1995; McCauley et al., 1997; Stein et al., 1996). The little published English-language data available from non-industrialized counties have shown that CSA is also a problem in these settings. Krugman, Mata, and Krugman (1992) found that 32% of university women in Costa Rica reported a history of CSA. A non-published study of 200 university women in Pakistan by the organization Daste-Shafqat in 1992 similarly showed 26% reporting a history of CSA. Singh, Yiing, and Nurani (1996) found a lower prevalence of 8.3% among female nursing and medical assistant trainees in Malaysia. Published cases of conrmed or suspected sexual abuse have been reported in Brazil (Farinatti, Fonseca, Dondonis, & Brugger, 1990), El Salvador (Rodriguez & Loza, 1991), India (Segal & Ashtekar, 1994), and Malawi (Lema, 1997). While the case reports included children from poor or rural areas, the systematic studies drew from presumably wealthier and more educated populations. Studies of sequelae have not been conducted in non-industrialized countries, to our knowledge. The purpose of this study, therefore, was to examine child sexual abuse among women in a poor, rural community in El Salvador, a developing country. The goals were to examine the prevalence and characteristics of CSA in this community and to determine whether a history of CSA was associated with psychological symptoms, taking into consideration possible trauma associated with the recent civil war.

METHODOLOGY Setting of Study The survey was conducted in a small, rural community in central El Salvador. El Salvador suffers many of the problems common to developing nations: poverty, high infant mortality, and low literacy rates. In addition, like several of its Central American neighbors, the country is recovering from a civil war that lasted throughout the 1980s into the early 1990s. Land ownership was a major issue in the early stages of the Salvadoran war. The study community, which is located approximately 3 hours outside of the capital, San Salvador, was established as an agricultural cooperative community in 1984 during the rst phase of agrarian reform. A few of the residents had lived there prior to the resettlement, but 93% had come from other parts of El Salvador. The community was poor; there was no electricity or telephone. Two spigots and a stream supplied the entire community with water. Average yearly income was difcult to ascertain, but families often spoke of difculty meeting food requirements. Housing consisted of mud units of one to three rooms. Although the fact that this is a land reform community makes it somewhat atypical, many Salvadorans were displaced or moved from their homes during the war. Thus, the experience of the communitys inhabitants mirrors that of many citizens. Social and economic conditions were similar to that found throughout El Salvador.

Sexual abuse in El Salvador

1119

Within families, women generally tended to the house, cooking, and children. Most men worked in the elds. A few men had positions in other communities or in the city. The Instrument The survey was administered as an interview. Questions were translated from English to Spanish by a local translator. The survey was pretested, with Salvadoran women asked to explain how they understood the questions. Wording was adjusted as required. The survey consisted of four parts: I. General demographic information, including questions about the subjects age, occupation, marital status, number of people in the household, education level, and literacy. II. History of child sexual abuse. Questions from the 1985 LA Times Sexual Abuse Survey (Finkelhor, Hotaling, Lewis, & Smith, 1990) were used to identify victims of child sexual abuse. The four questions are: 1. When you were a child (age 18 or under), can you remember having any sexual experience you would now consider sexual abuselike someone trying or succeeding in having any kind of sexual intercourse with you, or anything like that? 2. When you were a child, can you remember any kind of experience that you would now consider sexual abuse involving someone touching you, or grabbing you, or kissing you, or rubbing up against your body either in a public place or privateanything like that? 3. When you were a child, can you remember any kind of experience that you would now consider sexual abuse involving someone taking nude photographs of you, or someone exhibiting parts of their body to you, or someone performing some sex act in your presence or anything like that? 4. When you were a child, can you remember any kind of experience that you would now consider sexual abuse involving oral sex or sodomy or anything like that? The rst question was modied to differentiate further victims of completed versus attempted intercourse. Following positive responses to any of these four screening questions, data were obtained about age at time of abuse, perpetrators, and type and duration of abuse. III. Experience with war. Two questions addressed respondents experiences with the war: 1. Did you participate in the war? 2. Did you have any relatives killed during the war. If so, how many? IV. Screening of symptoms. The Hopkins Symptom Checklist (Derogatis, Lipman, Rickels, Uhlenhuth, & Covi, 1974) was used to examine the effects of CSA. The checklist screens for ve clusters of psychological symptoms: somatization, obsessive-compulsive behavior, interpersonal sensitivity, depression, and anxiety. Subjects rate the frequency with which they have experienced each symptom. A higher score indicates more pathology. A 5-item checklist for dissociative symptoms (Briere & Runtz, 1988) was included initially. A pilot study, however, showed that despite pretesting, certain concepts such as feeling outside of your body and not feeling like your real self were poorly understood and elicited confusion among interviewees. Those questions were then omitted. Study Population The community had 91 eligible women, operationally dened as those 18 years or older. Of these women, six were not at home and two refused. Eighty-three women participated in the survey, giving a response rate of 91%. Interviews were conducted by the rst author with assistance of one of two local female translators, who spoke directly with the women, and were familiar with the community, but did not live in it. The translators worked with the rst author to ensure a consistent approach and wording of interview questions.

1120

L. M. Barthauer and L. M. Leventhal Table 1. Percentage of All Women Surveyed Who Reported Each Type of Abuse in El Salvador, Costa Rica, and the United States El Salvador (n 83) Attempted/Completed Intercourse Touching/Grabbing/Kissing Exhibitionism/Photos Oral Sex/Sodomy
a

Costa Ricaa 10.8 (24/222) 29.0 (62/214) 5.2 (11/212) 2.4 (5/208)

United States (n 1481) 14.6 19.6 3.7 0.1

9.6 8.4 3.6 0.0

Demoninator varies for each type of abuse.

Interviews ranged in length from 20 minutes to 3 hours. The typical interview lasted approximately 1 hour. Shorter interviews were associated with those women who did not receive the HSCL. Longer interviews generally resulted from women discussing life in the community and details of the various symptoms listed in the HSCL. The purpose of the study was explained to all participants. Agreement to participate implied consent. Data about demographics, sexual abuse, and war experience were gathered on all women. All subjects with a positive history of sexual abuse were administered the Hopkins Symptom Checklist (HCL). Of women with negative histories of sexual abuse, every third subject was given the HCL. Statistics Statistical analysis was conducted using the SAS software package. Statistical signicance was specied as p .05.

RESULTS Demographic Information Women ranged in age from 18 to 70 years, with a mean of 34.7 years. All of the women considered themselves housewives; 2% sold small goods in their homes, as well. The women had an average of 4.7 living children (range 0 to 11) and an average of .6 children who had died (range 1 to 5). Households consisted of an average of 5.6 people (range 1 to 17). Nearly half of the women (49.4%) had no formal education. A third (36.2%) had attended school for one to two years. More than half (54.2%) considered themselves illiterate. Approximately one-third stated that could read and write a little. Only 9.6% considered themselves literate. History of Sexual Abuse Of the 83 participants, 14 (17%) reported a total of 21 experiences of sexual abuse. There was no statistically signicant difference in age between those reporting a positive history (mean 35.2) and those reporting a negative history (mean 32.1). Of the 21 experiences, seven (33.3%) involved attempted intercourse; three (14.3%), completed intercourse; eight (38.1%), grabbing/touching/ kissing; and three (14.3%), exhibition. As shown in Table 1, eight women (9.6%) reported at least one episode of attempted/completed intercourse, seven (8.4%) reported at least one episode of

Sexual abuse in El Salvador Table 2. Means of HSCL Scores Abused (n 14) Somatization Obsessive-Compulsive Interpersonal Sensitivity Depression Anxiety 26.8 15.0 13.4 24.6 12.1 Non-Abused (n 24) 25.3 14.2 11.6 21.4 12.5 p 0.43 0.50 0.08 0.04* 0.72

1121

*This difference did not reach statistical signicance after controlling for the number of relatives killed during the war.

touching/kissing/grabbing, and three (3.6%) reported at least one episode of exhibition. There were no reported cases of oral sex or sodomy. The mean age at the time of abuse was 13.9 years (median was 14 years). Seven events involved strangers, 12 involved a friend or neighbor, and two involved relatives. Of note, all perpetrators were male, and none was a parent or step-parent. Most of the reported abuse (16 of the 21 experiences, or 76%) occurred as a single event. One experience lasted less than 1 week, one lasted between 1 to 2 weeks, two lasted 2 months, and one lasted more than 5 years. Nearly all women (86%) were living with both parents at the time of the abuse. War Experience While many of the women reported providing soldiers or guerrillas with food, none reported active participation in the civil war. Slightly more than half of the women (52%) had at least one relative killed in war-related activity: 16% lost one relative to the war, 17% lost two relatives, and 19% lost three or more relatives. Symptoms Thirty-eight women completed the HSCL: 14 abused and 24 non-abused. Table 2 shows the means for the symptom clusters for the two groups. The only signicant difference between abused and non-abused women was on the depression measure, where abused women scored signicantly higher (p .04). When linear regression was used to control for the number of relatives killed during the war, however, this difference in depression between groups did not achieve statistical signicance (p .10). One-way analysis of variance (ANOVA) showed no signicant differences on any of the symptom clusters between non-abused women, women who reported attempted or completed intercourse, and those who reported any other form of CSA. DISCUSSION History of Sexual Abuse The rst goal of this study was to examine the epidemiologic characteristics of CSA in a poor, rural Salvadoran community. To our knowledge, this is the rst community-based survey in Latin America. The prevalence of 17% falls within the range described in Finkelhors (1994) review of the international epidemiology of CSA. The studies in his review, however, used many different methodologies and denitions of CSA. A more appropriate comparison would be with studies that used a similar screening tool as used in El Salvador. The LA Times sexual abuse questions have been used by Finkelhor, Hotaling, Lewis, and Smith (1990) and Krugman, Mata, and Krugman

1122

L. M. Barthauer and L. M. Leventhal Table 3. Prole of Perpetrators (of Reported Abuse, Percentages) El Salvador Stranger Friend/neighbor Parent Step-parent Other relative Other 33.3 57.1 0.0 0.0 9.5 0.0 Costa Rica 23.2 25.3 5.3 6.3 30.5 9.5 United States 21 41 3 3 23 9

(1992). In the original LA Times survey, Finkelhor, Hotaling, Lewis, and Smith found a prevalence of 27% in a telephone survey in the United States. Krugman, Mata, and Krugman found a prevalence of 32.2% when the same questions were given in a self-administered questionnaire to college students in Costa Rica. Thus, the prevalence of reported CSA in the community in El Salvador appears lower than described in comparable studies. The characteristics of abuse in this community also differ from the other studies. The median age at the time of abuse (13.9 years) is notably older than that described by Finkelhor and colleagues (1990) (9.6 years) and Krugman, Mata, and Krugman (1992) (8.0 years). The types of abuse differ as well, as shown in Table 1, with a smaller percentage of women in El Salvador reporting touching/kissing/grabbing or oral/anal sex than in Costa Rica or the United States. The proles for perpetrators in the three surveys are shown in Table 3. Most marked is the low percentage of relatives, the absence of fathers and step-fathers, and the high percentage of strangers, friends, and neighbors in the community in El Salvador. Duration of abuse was similar to that found in other studies, with single events constituting 65.2% of reported abuse in Costa Rica and 64% in the LA Times survey. In summary, compared to Costa Rica and the United States, victims of CSA in this community tended to be older at the time of abuse and were more likely to experience attempted or completed intercourse than to experience other forms of abuse. In addition, the perpetrators were likely to be strangers or acquaintances rather than relatives. Several possible explanations exist for these differences. First, the family and social structures in the rural community in El Salvador are much different than those in developed countries and in urban areas of developing countries. Children in the studied community are under almost constant supervision of mothers, grandmothers, and female neighbors. Until adolescence, when girls leave the home to work, there is little opportunity for them to be left alone, especially with men. Moreover, the large number of people in each household, the openness of the homes, and the close proximity of homes to each other, allow for little privacy. At puberty, when girls begin to leave their homes and to be alone more, they become more at risk of abuse by both acquaintances and strangers. The higher proportion of attempted and completed intercourse, compared to other forms of abuse, may be secondary to the type of perpetrator. In a study of records of sexually abused children, DeJong, Hervada, and Emmett (1983) found a greater risk of violence associated with assaults by strangers. The absence of abuse by relatives, especially fathers, is more difcult to explain. Virginity prior to commitment to a man is prized in this society. It is possible that fathers did not want to jeopardize their daughters opportunities for marriage. In other societies where virginity is equally important, such as in Asia and the Middle East, however, incest still occurs (Ho & Kwok, 1991; Roth, 1993). A study of 471 female Malaysian paramedical students (Singh et al., 1996) also found no reports of abuse by fathers. The authors suggest that in the population they studied, fathers often are dominant over daughters, even in adulthood, thereby limiting daughters disclosures. A similar dynamic might be operating in our population. A second explanation for the differences in CSA in this community is non-disclosure by the

Sexual abuse in El Salvador

1123

women, as mentioned above. Privacy was difcult to obtain during interviews. Houses were relatively open with no rmly closing doors. Women, therefore, may have been fearful of being overheard by neighbors. In two cases, husbands refused to leave during interviews. One of these women reported a positive history, nonetheless. Given the importance of pre-marital virginity, women may have been afraid of reporting past abuse for fear that the information would reach their current partners. Finally, some of the results, such as the absence of reports of oral and anal intercourse, may have been affected by the small sample size. The presence of an English-speaking foreigner at the interviews also should be considered as a factor affecting disclosure, both of CSA in general, and of familial abuse. While it would be difcult to ascertain the full effect of the rst authors presence, steps were taken to minimize it. First, a representative of the local womens group introduced and endorsed the project to the women on the rst day of interviews. Second, women were reassured that the information would be kept completely condential. Third, women were interviewed in settings of their choice to ensure comfort. The openness of many women in discussing CSA and other personal aspects of their seems to indicate that the presence of an English-speaking American did not adversely affect reporting; however, the possibility cannot be dismissed completely. Psychological Symptoms The second goal of the study was to examine the effect of CSA on psychological symptoms, taking into account the effects of the civil war. Our study showed no signicant differences between abused and non-abused women in psychological measures after controlling for the number of relatives killed in the war. In a meta-analysis of the literature on CSA and psychological adjustment, Jumper (1995) found that signicant relationships exist between a history of CSA and psychological symptomatology (including anxiety, personality disorders, suicidality, somatoform and dissociative disorders), depression, and self-esteem. Similarly, Briere and Runtz (1988) found a history of CSA to be related to somatization, anxiety and depression on the HCL in a study of college students. Using a different version of the HCL, however, Fromuth (1986) found no signicant differences between abused and non-abused college women on eight of nine measures after controlling for parental support. She attributed that lack of effect of CSA to a very broadly inclusive denition of sexual abuse, which did not differentiate between more invasive forms of CSA such as intercourse and less invasive forms such as exhibitionism. The current study showed no signicant difference on symptoms between victims of attempted or completed intercourse, and those experiencing less invasive forms of CSA like exhibitionism. The lack of relationship between CSA and psychological functioning in our sample may be a reection of other aspects of the womens lives, both in adulthood and in childhood. As adults, all of the women lived in poverty. Many had suffered the early death of a child. Although none of the women reported engaging in direct combat, the lives of most Salvadorans were touched by the decade-long civil war. One may postulate, therefore, that the cumulative effect of these experiences in the womens adult lives blunted the effects of the CSA. Additionally, while we did not ask about social conditions during childhood, such effects should not be overlooked. Most of the women in this community came from economic and social situations similar to the ones in which they are now living. While resettlement took place with intact families and was non-coercive, it likely represented a major life event. The younger women would have experienced political uprisings and war as children. These experiences, therefore, might have inuenced their adult psychological functioning as well. In support of this theory, Walton, Nuttall, and Nuttall (1997) found in a study of Salvadoran children born into the war that the personal/social impact of war, as measured by number of moves and lack of clothing and housing, was important in determining the mental health status of children. In the United States, Paradise, Rose, Sleeper, and Nathanson (1994) found in a group of school children with a history

1124

L. M. Barthauer and L. M. Leventhal

of sexual abuse that pre-existing psychosocial circumstances, rather than CSA, affected childrens behavior and psychiatric status. A study of women with a past history of child sexual abuse in New Zealand by Mullen, Martin, Anderson, Romans, and Herbison (1993) also found that some adult psychopathology could be accounted for by disturbed childhood, family, and social backgrounds. Thus, signicant problems in both childhood and adulthood may contribute to adult psychopathology, perhaps masking the effects of CSA. Another possibility for the lack of signicant relationship between psychological symptoms and a history of CSA is the predominance of abuse by strangers and acquaintances, as opposed to abuse by relatives. Briere and Runtz (1988) found that parental incest was related to somatization and anxiety. The effects of abuse by a stranger may be less traumatic than abuse by a parent. Additionally, the older mean age at time of abuse may be related to less psychological impairment, as increased disturbance has been associated with the early age of onset of abuse (Nash, Zivney, & Hulsey, 1993). A nal question raised by our results is whether Latin American women react differently to the experience of CSA. By virtue of their past experiences with poverty and conict, the women in this study do not necessarily represent typical Latin American women, and comparisons should be made with much caution. Studies in the United States examining differences in psychological symptoms between Latina and other women, however, have found no signicant ethnic differences (Mennen, 1995; Rao, DiClemente, & Ponton, 1992). While these studies included women no longer living in their native countries and presumably inuenced by lifestyles and values in the United States, the data suggest that ethnicity would not account for the lack of effect of CSA in the women in this study. More research is needed in developing countries to understand fully the epidemiology and effects of CSA on women. Careful attention should be paid to language and interview settings. While this study showed no difference in symptoms between abused and non-abused women after controlling for number of relatives killed in the war, the interviewees descriptions of abuse could be quite emotional, and indicated that CSA is indeed an important issue for these women. Future studies should seek to clarify the inuence of social factors on womens psychological adjustment and the interaction between CSA and social conditions.

REFERENCES
Bendixen, M., Muus, K., & Schei, B. (1994). The impact of child sexual abusea study of a random sample of Norwegian students. Child Abuse & Neglect, 18, 837 847. Briere, J., & Runtz, M. (1988). Symptomology associated with childhood sexual victimization in a nonclinical adult sample. Child Abuse & Neglect, 12, 5159. Browne, A., & Finkelhor, C. (1986). Impact of child sexual abuse: A review of the research. Psychological Bulletin, 99, 66 76. Cheasty, M., Clare, A. W., & Collins, C. (1998). Relationship between sexual abuse in childhood and adult depression: Case control study. British Medical Journal, 316, 198 201. Derogatis, L. R., Lipman, R. S., Rickels, K., Uhlenhuth, E. H., & Covi, L. (1974). The Hopkins symptom checklist (HCL): A self-report symptom inventory. Behavioral Science, 19, 115. DeJong, A. R., Hervada, A. R., & Emmett, G. A. (1983). Epidemiologic variations in childhood sexual abuse. Child Abuse & Neglect, 7, 155162. Farinatti, F. A. S., Fonseca, N. M., Dondonis, M., & Brugger, E. (1990). Child abuse and neglect in a developing country. Child Abuse & Neglect, 14, 133134. Finkelhor, D., & Korbin, J. (1988). Child abuse as an international issue. Child Abuse & Neglect, 12, 323. Finkelhor, D., Hotaling, G., Lewis, I. A., & Smith, C. (1990). Sexual abuse in a national survey of adult men and women: Prevalence, characteristics, and risk factors. Child Abuse & Neglect, 14, 19 28. Finkelhor, D. (1994).The international epidemiology of child sexual abuse. Child Abuse & Neglect, 18, 409 417. Fromuth, M. E. (1986). The relationship of childhood sexual abuse with later psychological and sexual adjustment in a sample of college women. Child Abuse & Neglect, 10, 515. Goldman, J., & Padayachi, U. (1997). The prevalence and nature of child sexual abuse in Queensland, Australia. Child Abuse & Neglect, 21, 489 498.

Sexual abuse in El Salvador

1125

Halperin, D. S., Bouvier, P., Jaffe, P. D., Mounoud, R., Pawlak, C. H., Laederach, J., Wicky, H. R., & Astie, F. (1996). Prevalence of child sexual abuse among adolescents in Geneva: Results of a cross-sectional survey. British Medical Journal, 312, 326 329. Ho, T., & Kwok, W. (1991). Child sexual abuse in Hong Kong. Child Abuse & Neglect, 15, 597 600. Jumper, S. A. (1995). A meta-analysis of the relationship of child sexual abuse to adult psychological adjustment. Child Abuse & Neglect, 19, 715728. Krugman, S., Mata, L., & Krugman, R. (1992). Sexual abuse and corporal punishment during childhood: A pilot retrospective survey of university students in Costa Rica. Pediatrics, 90, 157161. Lema, V. M. (1997). Sexual abuse of minors: Emerging medical and social problem in Malawi. East African Medical Journal, 74, 743746. MacMillan, H., Fleming, J., Trocme, N., Boyle, M., Wong, M., Racine, Y., Beardslee, W., & Offord, D. (1997). Prevalence of child physical abuse and sexual abuse in the community: Results from the Ontario Health Supplement. Journal of the American Medical Association, 278, 131135. McCauley, J., Kern, D., Kolodner, K., Dill, L., Schroeder, A., DeChant, H., Ryden, J., Derogatis, L., & Bass, E. (1997). Clinical characteristics of women with a history of childhood abuse: Unhealed wounds. Journal of the American Medical Association, 277, 13621368. Mennen, F. (1995). The relationship of race/ethnicity to symptoms in childhood sexual abuse. Child Abuse & Neglect, 19, 115124. Mullen, P. E., Martin, J. L., Anderson, J. C., Romans, S. E., & Herbison, G. P. (1993). Childhood sexual abuse and mental health in adult life. British Journal of Psychiatry, 163, 721732. Nash, M. R., Zivney, O. A., & Hulsey, T. (1993). Characteristics of sexual abuse associated with greater psychological impairment among children. Child Abuse & Neglect, 17, 401 408. Paradise, J. E., Rose, L., Sleeper, L. A., & Nathanson, M. (1994). Behavior, family function, school performance, and predictors of persistent disturbance in sexually abused children. Pediatrics, 93, 452 459. Rao, K., DiClemente, R. J., & Ponton, L. E. (1992). Child sexual abuse of Asians compared with other populations. Journal of the American Academy of Child and Adolescent Psychiatry, 31, 880 886. Rodriguez, M. E., & Loza, A. C. (1991). Records of Hospital de Ninos, Benjamin Bloom. San Salvador, El Salvador. Roth, K. (1993). Islamists: My rst six months. Newsletter of the Institute of Current World Affairs. Hanover, NH: Institute of Current World Affairs. Segal, U., & Ashtekar, A. (1994). Detection of intrafamilial child abuse: children at intake at a childrens observation home in India. Child Abuse & Neglect, 18, 957967. Singh, HSS A., Yiing, W. W., & Nurani, H. N. K. (1996). Prevalence of childhood sexaul abuse among Malaysian paramedical students. Child Abuse & Neglect, 20, 487 492. Stein, M. B., Walker, J. R., Anderson, G., Hazen, A. L., Ross, C. A., Eldridge, G., & Forde, D. L. (1996). Childhood physical and sexual abuse in patients with anxiety disorders and in a community sample. American Journal of Psychiatry, 153, 275277. Tonmyr, L. (1998). International studies on the incidence and prevalence of child maltreatment: Selected bibliography. Ottawa, Canada: Child Maltreatment Division, Bureau of Reproductive and Child Health, Health Protection Branch, Health Canada. Walton, J., Nuttall, R., & Nuttall, E. (1997). The impact of war on the mental health of children: A Salvadoran study. Child Abuse & Neglect, 21, 737749.

RESUME
Objectif: Cette etude retrospective a voulu examiner les aspects epidemiologiques des mauvais traitements sexuels chez des femmes vivant dans une collectivite pauvre et rurale du Salvador, pays qui tente de se remettre dune guerre civile qui a dure 12 ans. Methode: On a fait un sondage de prote en porte aupres de 83 femmes, en se servant du LA Times Sexual Abuse Survey ` pour preciser la prevalence des abus sexuels en enfance et pour mesurer les sequelles psychologiques a long terme (soit la ` somatisation, langoisse, la depression, la sensibilite interpersonnelle et les traits obsessifs-compulsifs). On a aussi obtenu des renseignements sur leur participation a la guerre et le nombre de membre de leur famille qui ont trouve la mort durant ` la guerre civile. Resultats: Parmi les participantes au sondage, 14 (17 p.c.) Ont rapporte 21 experiences dabus sexuels en enfance. Lage moyen etait de 14 ans. La majorite des agresseurs (90, 4 p.c.) etaient des etrangers, des amis ou des voisins. Daucun netait un membre de la famille. La seule difference entre les femmes abusees et celles qui ne letaient pas est au niveau de la depression, laquelle etait plus elevee chez les femmes abusees. Toutefois, apres avoir pris en consideration le nombre de ` membres de la famille morts en guerre, on a constate peu de differences ayant une importance statistique. Les caracteris tiques epidemiologiques du Salvador par rapport aux abus sexuels en enfance different des autres pays en ce que ce pays ` rural connat une structure sociale particuliere, et les femmes ne sont pas portees a reveler les agressions. Le peu de ` ` differences entre les symptomes psychologiques des femmes abusees et celles qui ne le sont pas pourrait sexpliquer par la nature des agressions et des agresseurs. Il se pourrait que les conditions sociales telles que la pauvrete et la guerre aient masque les effets a long terme des abus sexuels. `

1126

L. M. Barthauer and L. M. Leventhal

RESUMEN
Objetivo: El proposito de este estudio retrospectivo fue examinar la epidemiologa del abuso sexual en la ninez (ASN) en jujeres de una comunidad pobre, rural en El Salvador, que se recuperaba de una guerra civil de 12 anos. Metodos: Se administro una encuesta transversal puerta a puerta a 83 mujeres. Se utilizo el LA Times Sexual Abuse Survey para determinar la prevalencia del ASN. Se utilizo el Hopkins Symptom Checklist (HCL) para medir las secuelas psicologicas a largo plazo del abuso (somatizacion, ansiedad, depresion, sensibilidad interpersonal, y rasgos obsesivo compulsivos). Se obtuvo tambien informacion acerca de su participacion y numero de parientes muertos en la guerra civil de 12 anos en su pas. Resultados: De las 83 mujeres entrevistadas, 14 (17%) reportaron un total de 21 experiencias de ASN. La media de la edad del abuso fue 14 anos. La mayora de los perpetradores (90.4%) eran extranos, amigos o vecinos. Ninguno era padre o madre. La unica diferencia signicativa entre las mujeres abusadas y las no abusadas fue en la medida de depresion, en que las mujeres abusadas mostraron mas patologa. Despues de controlar el numero de parientes muertos en la guerra, sin embargo, la diferencia n llego a tener signicacion estadstica. Conclusiones: Las diferencias en la epidemiologa del ASN de los otros pases puede ser secundaria a una estructura social diferente en la comunidad rural del Salvador y a la no denuncia por las mujeres entrevistadas. La ausencia de diferencia en los sntomas psicologicos entre las abusadas y las no abusadas puede estar relacionada con las diferentes caractersticas del abuso y los perpetradores. Puede tambien ser secundario a las condiciones sociales adversas como la pobreza y la guerra, que pueden oscurecer los efectos a largo plazo del ASN.

This article is being published without benet of authors review of the corrected proof as this was not available at press time.