Article Text

Download PDFPDF

National study of physical and sexual assault among women with disabilities
  1. C Casteel1,2,
  2. S L Martin2,3,
  3. J B Smith2,
  4. K K Gurka1,4,
  5. L L Kupper5
  1. 1
    Department of Epidemiology, University of North Carolina, Chapel Hill, NC, USA
  2. 2
    Injury Prevention Research Center, University of North Carolina, Chapel Hill, NC, USA
  3. 3
    Department of Maternal and Child Health, University of North Carolina, Chapel Hill, NC, USA
  4. 4
    Department of Public Health Science, University of Virginia, Charlottesville, VA, USA
  5. 5
    Department of Biostatistics, University of North Carolina, Chapel Hill, NC, USA
  1. Dr C Casteel, Injury Prevention Research Center, University of North Carolina, 137 E Franklin Street, Suite 500, Chapel Hill, NC 27599-7505, USA; ccasteel{at}email.unc.edu

Abstract

Objective: To examine the association between the level of disability impairment and physical and sexual assault in a sample of US women at least 18 years of age.

Design, setting and participants: Retrospective longitudinal study of 6273 non-institutionalized US women from 8000 women participating in the 1995–1996 National Violence Against Women (NVAW) Survey.

Main outcome measure: Women’s experiences of physical and sexual assault in the 12 months before the NVAW interview.

Results: Most women reported having no disability (n = 5008, 79.8%) and/or not experiencing an assault in the year before their interview (n = 6018, 95.9%). Less than 5% (n = 280) reported having a disability that severely limited daily activities, and 15.7% (n = 985) reported having a disability that moderately limited activities. Less than 4% (n = 218) of the women reported a physical-only assault, and less than 1% (n = 37) reported being sexually assaulted. Women with severe disability impairments were four times more likely to be sexually assaulted than women with no reported disabilities (RR = 4.0, 95% CI 1.5 to 10.6). Little difference in the risk of sexual assault was found between women with moderate disability impairments and those reporting no disabilities (RR = 1.0, 95% CI 0.3 to 2.8). Women with severe (RR = 1.6, 95% CI 0.9 to 3.0) and moderate (RR = 1.2, 95% CI 0.8 to 1.9) disability impairments were at greater risk, although not quite significantly so, of physical-only assault than were women without a disability.

Conclusion: The findings suggest that women with disabilities that severely limit activities of daily living are at increased risk of sexual assault.

View Full Text

Statistics from Altmetric.com

Women with disabilities are a group vulnerable to abuse. Their disabilities may limit their ability to defend themselves,13 report the abuse,2 or recognize inappropriate behaviors.14 Women with disabilities can also experience unique forms of abuse.5 As these women may depend on others to meet basic health and/or social needs, abuse can be perpetrated by taking away medical equipment needed for daily functioning (eg, wheelchair), withholding access to communication (eg, TTY machines), or refusing to prepare meals.5

A growing literature recognizes women with disabilities to be at increased risk of physical and sexual assault.610 Most of these studies used convenience samples of women with disabilities rather than nationally representative samples of the female adult population, did not include a comparison group of women without disabilities, and/or failed to control for variables potentially confounding the disability–abuse relationship. Four studies using population-based, statewide, or national surveys addressed these limitations. Two of the studies found that women experiencing intimate partner violence were more likely to report having a disability or activity limitation than women not reporting intimate partner violence,1112 and the remaining two found that women with disabilities were more likely to experience sexual assault than women without disabilities.1314

Few studies have examined whether the extent of functional impairment due to a disability is related to a woman’s risk of violence victimization,91115 and only one provides strong evidence of such a relationship.11 In that study, women often or sometimes limited by their disability were approximately twice as likely to experience intimate partner violence as women without a disability. This study was conducted in a representative sample of women in Canada; a similar study in US women has not yet been published.

The purpose of this research was to study a national sample of non-institutionalized US women to examine whether women with disabilities (including those with more severe and less severe disability impairments) are more likely than women without disabilities to have been victims of physical and sexual assault.

METHODS

Sample

The study uses data from the National Violence Against Women (NVAW) Survey, conducted from November 1995 to May 1996.16 The NVAW Survey, a joint project between the National Institute of Justice and Centers for Disease Control and Prevention, is a random-digit dialing sample of telephone households nationwide. The NVAW sample included 8000 women and 8000 men at least 18 years of age who were demographically similar to the general US population, with the exception of under-representation among older adults, African–Americans, Hispanic men, and those with less than a high school education.

This study used a restricted sample of 6273 women who had complete data for the variables of interest (ie, disability status, assault experiences, and confounding variables). Data were missing for 1727 (28%) of all eligible women (table 1). Of these women, a little over 2% (n = 38) did not respond to questions addressing disability status, and about 21% (n = 364) did not respond to questions addressing experiences of violence. Women excluded from the study did not differ significantly with regard to demographic and household characteristics from those included.

Table 1 Characteristics of the study sample, National Violence Against Women Survey, 1995–1996

Assessment

Disability status

A woman was considered to have a disability if she reported one of the following: (1) sustaining a serious injury, such as a spinal cord, neck, or head injury, that is disabling or interferes with normal activities; (2) having a chronic disease or health condition that is disabling or interferes with normal activities; (3) having a chronic mental health disease or condition, such as chronic depression or schizophrenia, that is disabling or interferes with normal activities. Furthermore, women were asked about the extent to which these disabilities interfered with normal activities during the past week, with scores ranging from 0  =  “not at all” to 5  =  “extremely”. We categorized women as not having a disability if their score was 0, as having a moderate disability for scores 1 through 3, and having a severe disability for scores 4 and 5. We included impairment in the definition of disability to correspond with the Americans with Disabilities Act, which states that a disability is “a mental or physical impairment that substantially limits one or more major life activities”.17

Experiences of violence

Physical assault experiences were assessed using 12 items from the NVAW Survey, which included physical contact with a weapon (eg, gun, knife), hands or feet, as well as threats with a weapon, by a boyfriend/girlfriend, stranger, or acquaintance. Participants were classified as having experienced physical assault if they reported that any of the 12 items occurred within the past year. Respondents’ sexual assault experiences were assessed using five items, which included attempted and/or completed forms of sexual contact by force and/or threat of force, by a boyfriend/girlfriend, date, stranger, or acquaintance. Participants were classified as having experienced sexual assault if they reported that any of the five items occurred within the past year. To reduce problems associated with recall, only assault experiences occurring over a 1-year period, rather than over the adult lifetime, were considered.

Women’s physical and sexual assault experiences were then grouped into one of three mutually exclusive categories: (1) no assault/violence experienced within the past year; (2) physical-only assault within the past year; (3) sexual assault with or without physical assault within the past year. Owing to the small number of women who experienced sexual assault only (n = 23), we did not exclude women reporting physical abuse from the sexual assault outcome measure.

Of the women reporting a disability, 82% reported having a disability before abuse occurring in the year before the interview, 13% reported having a disability within the same year of abuse, and timing could not be determined in 5% of the study population. Therefore, temporal ambiguity between the onset of disability and time of abuse is possible in 18% (n = 232) of the study population.

Variables treated as potential confounders included the respondent’s age (18–24, 25–34, 35–44, and 45+ years), race/ethnicity (other vs non-Hispanic white), employment status (employed full-time, part-time or other vs unemployed), marital status (divorced, separated, widowed or single vs married), education (less than high school graduate vs high school graduate or more), and insurance coverage (other vs Medicare or private), as well as the number of adults (more than respondent vs respondent only) and children (one or more vs none) in the home.

Statistical analysis

Descriptive analyses were conducted to examine the study sample. A multinomial logit model18 was used to examine women’s experiences of assault during the past year (ie, no assault, physical-only assault, sexual assault with or without physical assault) as a function of the women’s level of disability impairment (ie, no disability, moderately impaired, severely impaired), controlling for demographic and household characteristics. We used odds ratios from the logistic regression to estimate risk ratios; as the prevalence of the study outcomes was less than 5%, odds ratios should closely approximate risk ratios.19 All analyses were conducted using SAS V9.1 (SAS Institute Inc, Cary, North Carolina, USA).

RESULTS

Study population

Most women were at least 35 years of age (68%), non-Hispanic white (79%), employed (97%), married (64%), had at least a high school education (90%), and had private insurance and/or Medicare (79%) (table 1). Approximately 80% of the women lived with at least one other adult, and over half (53%) had children in the home.

Disability and experiences of violence

Most women reported no disability (80%), 16% reported moderate disability impairments, and 5% reported severe disability impairments (table 1). Of the women with a disability (n = 1265), 856 reported having a chronic disease or health condition, 449 reported having a serious injury, and 137 reported having a chronic mental health disease or condition (table 2).

Table 2 Frequency of disability types, overall and by type of assault, among women reporting a disability (n = 1265), National Violence Against Women Survey, 1995–1996

Nearly all women experienced neither physical-only nor sexual assault in the year before their NVAW Survey interview (96%, n = 6018) (table 1). Less than 5% (n = 218) experienced physical-only assault, and less than 1% (n = 37) experienced sexual assault (with or without physical assault). The most common disabilities reported by women who experienced a physical-only assault were back injury and depression (table 2). In women reporting a sexual assault, serious injuries were the most prevalent types of disability.

After controlling for demographic and household characteristics, women whose disabilities had a severe effect on daily living were four times more likely to be sexually assaulted than women without a disability (adjusted RR (aRR) = 4.0, 95% CI 1.5 to 10.6; table 3). Little difference in the risk of sexual assault was found between women reporting moderate disability impairments and women with no disabilities (aRR = 1.0, 95% CI 0.3 to 2.8). Women with severe disability impairments were 60% more likely to report a physical-only assault (aRR = 1.6, 95% CI 0.9 to 3.0), and women with moderate disability impairments were 20% more likely to report a physical-only assault (aRR = 1.2, 95% CI 0.8 to 1.9), compared with women without a disability, although these associations did not quite reach the traditional level of statistical significance.

Table 3 Estimated associations between disability impairment levels, assault status, and covariates based on multinomial logit regression (n = 6273), National Violence Against Women Survey, 1995–1996

DISCUSSION

We found that women with disabilities that severely limit activities of daily living are at greater risk of sexual assault than women without disabilities. This study is unique in that it was the first to examine disability impairment as a risk factor for violence in a US sample of women at least 18 years of age.

Key points

  • Research suggests that women with disabilities are at increased risk of physical and sexual assault; however, most studies have examined convenience samples and few have included comparison groups of women without disabilities.

  • Few studies have examined whether the extent of functional impairment due to a disability is related to a woman’s risk of violent victimization.

  • In a national sample of US women, women with disabilities that severely limit activities of daily living are at increased risk of sexual assault compared with women without disabilities.

Women reporting severe disability impairments were four times more likely to be sexually assaulted than women reporting no disabilities, which is similar to findings from a statewide North Carolina study.13 Only one national study examined the relationship between levels of disability impairment and violence against women.14 In that study, the outcome was restricted to intimate partner violence, and multiple types of abuse (physical, sexual, emotional, and financial) were combined in the analysis. Because the exposure and outcome variables in this study are defined differently from those used in our study, comparisons cannot be made.

We found that physical-only and sexual assault were reported by 3.5% and 0.6%, respectively, of the women in this study. These prevalence estimates are similar to those found in other national11 and statewide1213 studies of disability and abuse outcomes among adult women. The estimated prevalence of disability in these studies ranged from 17% to 26%, which is comparable to the percentage of women reporting a disability in this study (20%).

Although this study has many strengths, there are some limitations. Women with substantial cognitive deficits are probably excluded from the NVAW Survey because of their potential inabilities to participate in interviews. We found that 5% of the women in the study reported having a disability that severely limited activities of daily living. In the 1994–1995 Survey of Income and Program Participation (SIPP),20 ∼13% of all women (with a similar age distribution to the population in this study) were estimated to have a severe disability. Although the definitions of severe disability in the SIPP and this study differ to some degree, it is likely that women with serious cognitive problems are under-represented in the NVAW Survey. In addition, the prevalence of victimization for both women with and without disabilities may also be under-reported in the NVAW Survey because of the sensitive nature of the topic.

Approximately one-quarter of the women in the NVAW Survey were excluded from this study because of missing data. However, demographic and household characteristics did not differ between those included and excluded. Although disability preceded assault in most of the study population, temporal direction between the onset of disability and occurrence of physical and/or sexual assault is uncertain in 18% of the study population.

REFERENCES

View Abstract

Footnotes

  • Funding: This study was funded in part by the University of North Carolina Injury Prevention Research Center through a grant from the National Center for Injury Prevention and Control, Centers for Disease Control and Prevention (Grant No R49/CE00196-01).

  • Competing interests: None.

  • Ethics approval: This study is an analysis of publicly available, de-identified secondary data and does not meet the definition of research involving human subjects requiring institutional review board review.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.