Although sexual abuse of people with intellectual disabilities has emerged as a major issue in the West over recent years, few researchers have examined this issue in Taiwan. The prevalence and current state of sexual abuse for people with intellectual disabilities in Taiwan were investigated here. Results of face-to-face interviews with 336 subjects revealed that the prevalence of sexual abuse for this population is 5.4%. This low prevalence of sexual abuse has certain cultural implications; most people with intellectual disabilities live in supported living settings and are, therefore, excluded from community life. Increased effort is needed to develop effective service programs for abuse prevention and strategies should be developed to improve the limitations in the present study.
Although sexual abuse of people with intellectual disabilities has been an emerging issue in the West in recent decades, few researchers have examined this issue in Taiwan. Western studies reveal a higher prevalence of sexual abuse for people with intellectual disabilities compared to those without intellectual disabilities (Kempton & Gochros, 1986, as cited in Furey, Granfield, & Karan, 1994, p. 75; Sullivan & Knutson, 1994, as cited in Mansell, Sobsey, & Moskal, 1998, p. 12). Most researchers have investigated local or regional, rather than nationwide populations (Furey, 1994; Gust, Wang, Grot, Ransom, & Levine, 2003; Mansell et al., 1998; Valenti-Hein, 2002; Zirpoli, Snell, & Loyd, 2001). Moreover, most of these studies have been agency- rather than community-based (Furey, 1994; Gust et al., 2003; Mansell et al., 1998; Valenti-Hein, 2002; Zirpoli et al., 2001). For example, a retrospective study of in-patients with intellectual disability aged 9 to 21 years hospitalized over a 5-year period indicated that 14% had been sexually abused (Balogh et al., 2001). In a study of data from the Connecticut Office of Protection and Advocacy for Persons with Disabilities over a 5-year period, Furey (1994) found that 37% of people with intellectual disabilities had been sexually abused. Data for similar studies have been collected from service records (Balogh et al., 2001; McConkey & Ryan, 2001) or service providers (Firth et al., 2001; Furey et al., 1994). However, few investigators have analyzed data acquired from family caregivers or individuals with intellectual disabilities.
Current explanations for the phenomenon of sexual abuse for people with intellectual disabilities vary widely. Some investigators focus on individual characteristics, such as cognitive ability, life-long dependence on caregivers, and increased trust of strangers (Sobsey & Mansell, 1990; Strickler, 2001). Others point to institutional factors, such as insufficient supportive environments and coping resources, relative powerlessness in society, emotional and social insecurity, and lack of sex education. The relationship between different functional levels of intellectual disability and prevalence of sexual abuse remains unclear. Some investigators have suggested that mild or moderate disability increases the risk of sexual abuse (Liou, 2006; Martin, 1982), whereas others reported that severe disability increases the risk of abuse (Zirpoli et al., 2001).
Most studies demonstrate that perpetrators of sexual abuse are often known to the victims (Furey, 1994; Furey et al., 1994; Khemka & Hickson, 2000; Liou, 2006). In a 1994 study of sexual abuse cases in Connecticut, Furey reported that 71% of perpetrators were known by the victims, who were other individuals with intellectual disabilities, friends, family members, and neighbors; only 21% of perpetrators were strangers, such as service providers. This finding was further confirmed in a study by Liou (2006), who analyzed 164 case records of sexual abuse victims with intellectual disabilities in Taiwan from 1999 until 2006. All previous researchers have noted that females with intellectual disabilities are at increased risk for sexual abuse (Balogh et al., 2001; Furey, 1994; Liou, 2006; Mansell et al., 1998; Strickler, 2001).
According to Furey (1994), the percentage of female sexual abuse victims is as high as 95%, and the average age of sexually abused adults is 30 years; however, Liou (2006) reported an average age as low as 20.7 years. According to the prevailing stereotype, people with intellectual disabilities are often oversexed or asexual (Carlson, 1997; H. Chen & Liou, 2006; Furey, 1994; Mansell et al., 1998; Tharinger, Horton, & Millea, 1990). Therefore, in this study the current status and prevalence of sexual abuse in the population with intellectual disabilities were studied in Taiwan from a cross-cultural perspective based on information obtained directly from family caregivers or adults with mild intellectual disabilities.
According to third quarter statistics (2005) from the Ministry of the Interior, the population of people with intellectual disabilities in Taiwan was approximately 80,918, which became the study population. I selected participants from this population by multistage cluster sampling. First, 25 counties with 382 towns nationwide were divided into seven districts based on degree of urbanization. Second, I randomly selected 26 towns from seven districts according to percentages of subjects in each district. I also considered gender because it significantly influences the prevalence of sexual abuse. Finally, I randomly sampled a percentage of participants from each town. For budgetary reasons and for the convenience of interviewers, the number of randomly sampled participants was limited to 13 or 16. In total, 384 subjects were randomly selected from 26 towns. Sampling error was 5% at a 95% confidence level.
At completion of data collection from 343 participants (89.3% response rate), I excluded 7 participants due to incomplete answers. Therefore, the final study population consisted of 336 subjects, 146 females (43.5%) and 190 males (56.5%). The age distribution was as follows: 0–17, 88 (26.2%); 18– 44, 196 (58.3%); 45–64, 47 (14.0%); and 65 and above, 5 (1.5%) participants. Caregivers were also included (n = 232), 13 of whom had mild intellectual disabilities. Most people with intellectual disabilities in Taiwan used to lived with their family members. However, upon the death of their families, they tended to live alone. To accurately estimate the population distribution of people with intellectual disabilities in this environment, I also included 13 adults with mild intellectual disabilities living alone. Chi-square analysis of the sample populations by gender, age, and district yielded no significant differences; therefore, the analytical results of this study can be validly generalized to the overall population with intellectual disabilities in Taiwan (see Table 1).
In this study, sexual abuse was defined as inappropriately touching a victim or forcing, threatening, or coaxing a victim to perform a sexual act (Furey et al., 1994, p. 76). The questionnaire utilized consisted of three sections: basic information, experience of sexual abuse, and posttraumatic symptoms and utilization of social services. I modified the questionnaire after consultation with experts in this area. The final version had 38 questions. The first section, basic information, included 21 items that covered demographic information about the participants and their families, including gender, age, functional level, education level, marital status, employment status, family structure, number of family members, income level, and welfare aid. The second section had 12 questions about experiences of sexual abuse, including location of incidents, relationship between perpetrator and victim, and posttraumatic syndrome disorders (PTSD). Because the participants in this study were largely caregivers, questions regarding PTSD were designed as checklists of symptoms indicating psychological, physical, and adaptive behaviors. Information on physical symptoms included fatigue, loss of appetite, body ache, headache, menstrual pain, bedwetting, and insomnia. Information regarding psychological syndromes included absent-mindedness, a sense of insecurity, feelings of loneliness, deep anxiety, indifference, irritability, depression, guilt, reduced attention span, memory loss, and fantasizing. Information regarding adaptive behaviors included items about telling lies, running away from home, cutting class, vagrancy, stealing, aggression, inappropriate sexual behavior, drug abuse, alcohol use, and smoking. These items were included on the checklist. Before completing the questionnaire, participants were asked to identify any syndromes they had been diagnosed with. The final section, consisting of 5 items, was related to use of social services and needs of victims and their families. Because each item in the three sections was designed to be close-ended, rather than testing reliabilities I conducted a pilot study with 5 subjects to determine whether the terminology of each sentence was appropriate or not.
I conducted face-to-face interviews to gather information from caregivers or from adults with mild retardation who were living alone. Interviewers were social workers from local governments and nonprofit organizations. Each interviewer was required to complete a half-day training course. Interviews were scheduled by telephone, and a letter of introduction was sent after each interview was scheduled. Participants who could not be contacted by phone were visited at the address listed in the database of people with intellectual disabilities. Subjects were excluded if they had relocated elsewhere, could not be reached after three visits, or if they refused to participate. The replacement sample was 159 individuals.
All interviews were conducted in the participants' home. After explaining the study purpose, all interviewers requested oral consent before beginning the interview. Data were collected between September 13 and November 23, 2005. Of the 336 subjects, 314 (93.4%) were family caregivers, 13 (3.9%) were adults with mild disability who lived alone, and 9 (2.7%) were others, such as friends and neighbors. Descriptive statistics, such as frequency and chi-square analysis, were applied to analyze current status and prevalence of sexual abuse of people with intellectual disabilities.
Prevalence and Victim Characteristics
Of the 336 subjects, 18 (5.4%) confirmed having experienced sexual abuse. Therefore, an extrapolated estimate is 4,406 sexual abuse victims. Prevalence of sexual abuse was 7.4% for females and 4.2% for males. Sexual abuse victims and nonvictims did not significantly differ in gender, intellectual disability level, education level, marital status, or incidence of sexual abuse. Although results of previous studies have indicated that females with intellectual disabilities are at increased risk for sexual abuse (Balogh et al., 2001; Furey, 1994; Liou, 2006; Mansell et al., 1998; Strickler, 2001), this correlation was not confirmed in this study.
Of the 18 victims, 10 (55.6%) were female, and 8 (44.4%) were male, indicating that although more women than men were sexual abuse victims, the difference was not statistically significant. Victims and nonvictims did not significantly differ in functional levels of intellectual disability. Prevalence of sexual abuse in persons with mild, moderate, severe, and profound intellectual disabilities was 9 (7.4%), 6 (5.7%), 3 (3%), and 0 (0%), respectively. Prevalence in unmarried subjects was 8.6%, slightly higher than that in married participants (5.2%); however, the difference was not statistically significant. In total, 302 of these individuals (89.9%) currently lived with their family, only 12 (3.6%) lived alone, 15 (4.5%) lived in institutions, and 7 (2.0%) resided with friends.
Family characteristics, such as education level of parents, income level, and family structure, did not significantly differ between victims and nonvictims. Although no statistically significant difference was noted in educational level of parents, 10.8%, 3.5%, and 6.0% of the fathers were college-, high school-, and primary school-educated, respectively. Of the 336 participants, 100 (29.8%) lived with an extended family, fewer than half (48.2%) lived with a nuclear family, and one tenth lived in a single-parent family. Of the 18 victims, 14 (77%) lived in either a nuclear or single family, and only 4 victims lived with an extended family. Despite the higher prevalence of sexual abuse in families with low (6.7%) and high (6.8%) incomes than in families with medium incomes (2.6%), the differences between these groups were not statistically significant.
Perpetrators and Location of Sexual Abuse
Participants were asked to describe their relationship with the perpetrator. Five of 188 would not answer the question. Clearly, perpetrators were known to most victims. For 13 victims who were sexually abused, 9 perpetrators (69.2%) were either family members, relatives, classmates, neighbors, or government agency staff. Three perpetrators (23.1%) were other than those listed above. Only one perpetrator (7.7%) was unknown to the victim. Table 2 lists locations of sexual abuse for all participants, which included the residence, workplace, school or community, and the residence of the perpetrator.
Table 3 is a summary of the analytical results for PTSD symptoms of victims. Roughly 50% of victims had such symptoms. Psychological symptoms included absent-mindedness, anxiety, depression, irritability, and loss of concentration. Less than 50% reported having physical symptoms and behavioral problems.
Cultural Implications of Prevalence of Sexual Abuse
The overall prevalence of sexual abuse of people with intellectual disabilities in Taiwan (5.4%) is considerably lower than that reported elsewhere (e.g., Balogh et al. (2001) and Furey (1994)); however, most Western investigators (Gust et al., 2003; Valenti-Hein, 2002; Zirpoli et al., 2001) analyzed data from service providers or case records rather than from family caregivers. This result has several cultural implications. Theoretically, the prevalence of sexual abuse reported by nonfamily individuals should be higher than that reported by family members. In Taiwan, if an individual reported being sexually abused by a family member, it would be viewed as dishonoring the family, and, therefore, the incidence of sexual abuse tends to be underreported. For example, C. Chen (2004) explored the social stigma of people with intellectual disabilities and their families by interviewing 9 caregivers (all mothers) and found that the Buddhist concept of retribution for sin contributed to the underreporting of the population with intellectual disabilities.
The low prevalence of sexual abuse of people with intellectual disability may also be affected by an ideological belief that the family is responsible for the care of family members, especially vulnerable members. Within the context of a conservative welfare ideology, the family, rather than government, is responsible for supporting and protecting sexual abuse victims. Parents are expected to care for their children with intellectual disabilities regardless of how old the children may be. When parents pass away, their children may or may not assume responsibility for family members with intellectual disabilities. In line with these ideological beliefs, most participants in this study (302, 89.9%) lived with their families. Only 12 (3.6%), 15 (4.5%), and 7 (2.0%) lived alone, in institutions, or with friends, respectively.
Gender differences in the prevalence of sexual abuse also reflect certain cultural beliefs. Females are much more likely to be victimized than males. In a 2006 study, Liou reviewed case records from mental health facilities throughout Taiwan and found that 95% of sexual abuse victims were females. However, the analytical result of the present study revealed no such gender difference, indicating that the risk of sexual abuse is similar for both males and females with intellectual disabilities. Clearly, sex education is a vital need in public schools. Training programs for sexual abuse prevention are also urgently needed.
Zirpoli et al. (2001) and Martin (1982) also identified an ambiguous relationship between mental functional level and incidence of sexual abuse. According to Martin, individuals with mild disabilities are at high risk of sexual abuse, whereas Zirpoli et al. indicated that those with severe and profound disabilities are at high risk. Martin's study is consistent with the recent findings of Liou (2006) that individuals with mild disabilities are more likely to suffer sexual abuse than are those with moderate, severe, or profound disabilities. A reasonable explanation is that individuals with mild disabilities have more opportunities to venture into the community alone and interact with people outside the family. However, lack of public education, limited facilities for individuals with intellectual disabilities, and the social stigma of having family members with intellectual disabilities increase the risk of sexual abuse for these individuals. This explanation is confirmed by the roughly 66% of cases occurring in the public sphere, such as in institutions, work places, and schools, whereas about 33% of cases occur in the homes of victims and perpetrators. Prevailing negative stereotypes of individuals with intellectual disabilities produce an unfriendly community environment. The resulting exclusion of these people limits their independence. Therefore, effective programs are needed to prevent abuse and improve community inclusion of people with intellectual disabilities.
Lack of Formal Supportive System
Prevailing stereotypes of people with intellectual disabilities tend to overlook posttraumatic syndromes resulting from sexual abuse; however, results of this study indicate that more than 50% of victims have varying degrees of posttraumatic stress. Dominated by a stereotypical public image of this population and a conservative welfare ideology, the social and human rights of these victims are largely ignored by policymakers. Due to poor professional knowledge and inadequate experience working with this segment of the population, service providers are reluctant to allocate service to victims and their families and, in some cases, even display inappropriate attitudes toward the family (Liou, 2006). Ideally, sexual abuse prevention policies would be developed in consideration of the sexual activity of this population. Further, direct-care workers must be trained to detect signs of sexual abuse and to communicate with these individuals and their families.
As an initial study of the prevalence of sexual abuse for the population with intellectual disabilities in Taiwan, the analytical results of this study have several important implications. However, some limitations of the study design must be addressed. The data for this study were collected in retrospective interviews requiring interviewees to recollect past experiences of sexual abuse. Because a history of sexual abuse would be seen as dishonor to the family, interviewees may have underreported these incidents. However, the prevalence rate is likely to accurately reflect reality in Taiwan where the family unit tends to protect members with intellectual disabilities; in turn, this not only lowers the reported prevalence, it also impedes the independence of this population. To address this dilemma, I strongly suggest that agency-based quantitative studies or studies with larger survey sample sizes be conducted.
Sexual abuse is seen as a taboo and a family dishonor in Taiwan. Because of the sensitivity of this issue, questionnaires must be carefully worded. For example, the term sexual abuse is replaced by inappropriate touching. Further, interviewees were largely family caregivers who may have been insensitive to incidence of sexual abuse in individuals with intellectual disabilities and were perhaps unable to correctly recall the age and frequency of sexual abuse incidents. An additional limitation of the study is that interviewed caregivers or friends may have been the perpetrators of sexual abuse, which would result in underreporting. In this study, therefore, I adopted a strategy of avoiding issues regarding age and frequency of sexual abuse in the questionnaires. This study was exploratory because no standard records of sexual abuse are available in Taiwan for comparison with this study. All questionnaire items in the present study were categorical, thus limiting statistical analysis of reliability.
The replacement rate of the sample was 159 (41.4%) in urban areas, which was higher than that in rural areas. Two of 26 towns did not complete the interview process due to unexpected circumstances (e.g., accidents occurred during the period of interviewing), and they did not to inform the researcher until completion of the study. A high replacement rate may limit generalization of the prevalence rate in this sample to the overall population of individuals with intellectual disabilities. In addition, the limited awareness about sexual abuse in the general population and the tendency to avoid professional assistance should be further analyzed to clarify the interaction between individuals and the abuse prevention system. The needs of families of people with intellectual disabilities and the experiences of direct-care workers must be explored to improve the sexual abuse prevention system and delivery of mental health services.
In summary, although the results of this study revealed no significant differences in personal and family characteristics, certain cultural and structural implications germane to this issue were explored; therefore, increased cultural sensitivity when addressing this issue is suggested. Finally, the current sexual abuse prevention system in Taiwan must be re-examined so that this population can be afforded the same political rights as the general population.
The research reported in this article is part of the project A Survey Research of Violence Against People With Disabilities, Victims of Family Violence and Sexual Assault, supported by a grant from the Ministry of the Interior of the Republic of China, Taiwan, from March to December 2005.
Shu-Man Pan, PhD, Professor, Graduate Institute of Social Work, National Taiwan Norman University, 162 Sec. 1, E. Ho-Ping Rd., Taipei, Taiwan. firstname.lastname@example.org