A Longitudinal Analysis of Disability-Related Interpersonal Violence and Some Implications for Violence Prevention Work

Abstract

This article explores the extent to which disabled individuals experience interpersonal violence due to victimization. Data on people injured by violence were collated directly from the accident and emergency units in hospitals. High frequency daily data were obtained from computerized records of 26 major accident and emergency departments in London for each day throughout the year of 2016. The final sample consisted of 408,000 observations. A fundamental distinction of our research lies in applying the Generalized Method of Moments system panel estimator to our sample. This makes our empirical estimates robust to endogeneity and joint determination unlike previous empirical research in this area. Data analysis provides strong evidence confirming the victimization of people with disabilities and the necessity to focus on disability equality in violence prevention work.

Keywords

disability interpersonal violence human rights Generalized Method of Moments policy

Introduction

Despite the reconceptualization of disability as a human rights issue (United Nations, 2008), people with disabilities are still disproportionally victimized and experience violence (Emerson & Roulstone, 2014; Mikton & Shakespeare, 2014; Ralph et al., 2016). Violence has been manifested in varied forms and guises ranging from physical, sexual, psycho-emotional, cultural, and systemic forms, and experienced in mutually reinforcing and cumulative ways (Goodley & Runswick-Cole, 2011). Apart from their symbolic and psycho-emotional effects, violent acts result in injuries and chronic illnesses that adversely affect the quality of life of disabled individuals and raise serious concerns about public health policy constitution and dissemination (Marge, 2011; World Health Organization [WHO], 2011).

Given the importance of using the notion of “difference” as a means of exploring the ways in which the intersection of disabling social ecologies and practices precipitate violent and abusive behaviors linked to disability (e.g., Ralph et al., 2016), it is worth noting that issues of disability equality to prevent violence against people with disabilities seem to be absent from the WHO agenda. This is in direct contrast to an explicit focus on promoting gender equality to prevent violence against women (e.g., WHO, 2009).

A disability perspective in violence prevention work infers the imperative of promoting disability equality as a means of preventing violence. This can be achieved by addressing binary perspectives of normality and abnormality that legitimize the subordinate positioning of individuals with disabilities based on their presumed deviation from an arbitrarily fabricated “able-bodied order” (Campbell, 2009). Such an acknowledgment attests to the value of promoting an intersectionality-based policy analysis framework (IBPA) to challenge social hierarchies and structural inequalities that are accountable for interpersonal violence. An IBPA framework is explicitly concerned with asking transformative questions about the shortcomings of existing policies, and has the potential to facilitate policy changes and collaboration among diverse policy units (Clark, 2012; Hankivsky et al., 2012), with a view to dealing with issues of disability-related violence in effective and socially just ways.

These questions relate, for instance, to the extent to which education policies and practices address and prohibit disability stereotypes. Schools have a pervasive impact on shaping young people’s conceptualizations and encounters with “disability as difference” perspectives and constitute discursive spaces where “normality is learned” (McLaughlin, 2017, p. 59) and is responsible for breeding disability-related violence (Chatzitheochari, Parsons, & Platt, 2016).

In view of the above considerations, it is not surprising that a significant body of research provides consistent evidence to document strong associations between violence and disability (e.g., Harrell, 2016; Hughes et al., 2012; Hughes, Lund, Gabrielli, Powers, & Curry, 2011; Krnjacki, Emerson, Llewellyn, & Kavanagh, 2016; Marge, 2011). Other minority statuses linked to a disabled person’s racial, gendered, sexual, and classed characteristics have also been empirically reported to increase the risk of disability-related interpersonal violence (e.g., Olofsson et al., 2015; van der Heijden, Abrahams, & Harries, 2019).

Tackling disability-related interpersonal violence is a human rights issue (see Article 18, The United Nations Convention on the Rights of Persons with Disabilities [UNCRPD]) that needs to be prioritized and dealt with in systemic, evidence-based, and effective ways. Beyond rhetorical proclamations aligned with the rights-based stipulations of the UNCRPD, signatory nation states need to provide up-to-date empirically validated information on the ways in which disability rights are promoted and protected (see Article 35 of UNCRPD). The provision and constant updating of national data on disability-related violence are an imperative that needs to be constantly monitored and reviewed (Marge, 2011) to document progress and to formulate and implement intersectionality-based violence prevention policies and practices, which focus on public health and other policy implications of using a “disability perspective” in violence prevention work. As Mikton, Maguire, and Shakespeare (2014, p. 3210) point out, a public policy approach to violence against people with disabilities “begins with understanding the magnitude, distribution and consequences of the problem.”

The formulation of violence prevention policies and practices presuppose an understanding of the extent to which disabled individuals experience violence (Hughes et al., 2012), and an understanding of the intricate web of dynamics that put disabled individuals at a higher risk of experiencing social conflict and interpersonal violence (Marge, 2011).

Despite the availability of studies that try to establish links between violence and disability, Olofsson et al. (2015) point out that “even in high-income countries, data on prevalence and odds ratios for violence exposure for people with disabilities are lacking” (p. 1682). For instance, Hughes et al.’s (2012) meta-analysis of 26 studies that met eligibility criterial and sought to explore the relationship between violence and disability, their data were limited to 21,557 individuals with disabilities.

Apart from the dearth of adequate data on disability-related violence, existing studies are reported to have methodological weaknesses, while there is a lack of studies using high-quality population-based studies (Hughes et al., 2012; Marge, 2011; Mikton & Shakespeare, 2014; WHO, 2014). As far as the link between disability and physical violence is concerned, most of these studies could not distinguish whether disability was the result of violence or it constituted a risk factor for being subject to violence (Hughes et al., 2012; Mikton & Shakespeare, 2014). Simultaneously, Mikton and Shakespeare (2014, p. 3057) suggest that apart from the weak methodological quality of most studies, these studies do not provide data on specific types of disability and specific types of violence. Along similar lines, Hughes et al. (2012) suggest that these studies have gaps not only in the type of disability they address but also in the type of violence.

These are major weaknesses in trying to understand the extent to which individuals with disabilities are at higher risk of being subject to interpersonal violence. This information is crucial to develop a public health policy-making approach to preventing violence against disabled people (Hughes et al., 2012; Mikton et al., 2014) and, subsequently, contribute to the improvement of the material life circumstances experienced by individuals with disabilities (see Oliver, 2002).

Although our data do not contain information on the type of the self-reported disability, they contain observations based on 93,840 individuals with disabilities—this is a significantly larger database compared with prior studies (Hughes et al., 2012, Krnjacki et al., 2016; Khalifeh et al., 2013; Marge, 2011), and they provide evidence of actual inflictions of violence resulting to physical injuries reported by accident and emergency departments. This is significant because Mikton et al. (2014) and Hughes et al. (2012) point to the lack of studies using high-quality population-based studies to explore actual violence. For example, Mikton et al. (2014) conducted a meta-analysis of the effectiveness of interventions to prevent and to respond to violence against persons with disabilities using 10 studies which, according to their assessment criteria (see quality assessment tool for quantitative studies), these studies were considered “weak” as they measured risk factors associated with violence and not actual violence as an outcome. The latter could have been measured if these studies, according to Mikton et al. (2014, p. 3226) had data on actual violence that could be either ‘self-reported or collated from police, social work, or other official records.’

At the same time, our study overcomes the prevalent problem of high levels of nonresponse manifested in all population-based studies that cannot determine whether “non-respondents would have answered fairly similarly to those who did not answer” (Olofsson et al., 2015, p. 1682). Based on our data on actual violence, following up studies can evaluate the impact of violence prevention interventions in the light of new empirical evidence (Mikton et al., 2014). As the National Council on Disability (2007, p. 3) highlighted, “sound public policy, resource allocation, and program development must be informed by current, scientifically valid data”.

In this article, we also contribute to the previous research in the following ways. First, we are the only study to analyze the impact of the disability of an individual and actual physical violence on a daily basis. The longitudinal character of our research mitigates some of the shortcomings of cross-sectional designs that have been exclusively used to explore the prevalence of violence and relationship between impairment and violence (Olofsson et al., 2015). We do this by collecting data from accident and emergency units located in London on each day throughout the year of 2016. Preceding research was conducted on a cross-sectional level where you could not observe time series trends. We have a very large data set with 365 time periods, a total number of observations of 408,000 with 93,840 observations classified as disabled and 314,160 classified as nondisabled individuals. This is a significantly larger high frequency data set compared with prior studies (Hughes et al., 2012; Khalifeh et al., 2013; Krnjacki et al., 2016; Marge, 2011).

The second contribution of our study is on the methodology used to empirically examine the relationship between violence and the disability of an individual. Unlike prior research, we overcome contemporaneous correlation, endogeneity, and jointly determination of the violence injury rate and the disability of an individual by employing the Generalized Method of Moments (GMM) system panel estimator established by Blundell and Bond (1998) on our data. This makes our empirical estimates robust and, therefore, more reliable than the previous empirical research in this area.

The rest of the article is organized as follows. The following section discusses the econometric specification. The Data section discusses the data set; the Results section presents the empirical results. Finally the Discussion section summarizes and concludes.

Method

Econometric Specification

To conduct our empirical analysis, we model the relationship between the daily violent injury rate (V) and the disability of individuals by estimating the following equation:

V_{i t} = α_{i} + b_{t} + γ_{1} P_{i t} + γ_{2} D_{i} + ε_{i t}

(1)

where i represents the individuals in our sample and t denotes the daily time period; α captures the time-invariant unobserved violent injury rate individual-specific fixed effects (e.g., differences in the injury rate of individuals independent of disability, the age, and ethnicity of an individual), and the $b_{t}$ captures the unobservable individual-invariant time effects (e.g., changes in disability benefits that affect the association between the disability and the rate of violence for all individuals). The empirical relationship between disability and the violent injury rate of individuals is captured by the dummy variable D_i. The dummy takes a value of 1 if the individual suffers from a form of disability and 0 otherwise. We also include the price of alcohol as a control variable because Matthews, Shepard, and Sivarajasingham (2006) find a negative relationship between the price of alcohol and violence. P_it is the price of alcohol that each individual pays at each time period.

We also estimate two other versions of equation (1) to provide some insight into the causality of the association between disability and violence of individuals. We achieve this by distinguishing if the disabled individuals are the aggressors or the victims of the incident that results in the injury. This is accomplished econometrically by changing the definition of the dummy variable explained in the previous paragraph. In the first model, we capture if disabled individuals are the aggressors by assigning the value of 1 if the individual is the aggressor and 0 otherwise in the dummy variable. In the second model, we capture if disabled individuals are the victims by assigning the value of 1 if the individual is the victim and 0 otherwise in the dummy variable.

To formally test the explanatory variables for endogeneity, a Hausman (1978) test for the hypothesis that the explanatory variables are strictly exogenous is performed. If the null hypothesis is rejected, it leads to the conclusion that the explanatory variables in equation (1) are endogenously determined. The Hausman test rejects the null hypothesis at all conventional significance levels.¹

To accommodate endogeneity and the possibility of joint determination, we employ a GMM system of equations in first differences and levels to estimate equation (1).² The estimation of the systems of equations simultaneously using the GMM system should be (a) asymptotically efficient due to nonrestrictive assumptions about error autocorrelation and heteroscedasticity (Biørn & Klette, 1999), (b) accommodate the explanatory variables being jointly determined with individuals’ injury rates, (c) control possible relationships between the explanatory endogenous variables and the individual injury rates.

This system estimator combines the standard set of transformed equations in first differences (used in the GMM single equation estimator) with an additional set of equations in levels. The first set of transformed equations uses the lag levels as instruments and the level equation uses the lagged first differences as instruments. The first set of transformed equations continues to use the lag levels as instruments. The level equation, on the contrary, uses the lagged first differences as instruments. Their validity is based on the following two moment conditions:³

E [\begin{array}{l} (α_{i} + ε_{i t}) Δ V_{i, t - z} \\ (α_{i} + ε_{i t}) Δ X_{i, t - z} \end{array}] = 0 for z = 1

(2)

where $X_{i t}$ is a vector of the explanatory variables, and $z$ represents the lag structure in the econometric model.⁴

Data

We collected the data on people injured by violence directly from the hospitals. The data were gathered on a daily basis from computerized records for each day throughout the year of 2016. 26 major accident and emergency departments in London took part in our study.⁵ Each department has electronic records on whether the injury is caused by accident or by interpersonal violence, obtained from police records. Individuals self-report if they are disabled or not. Our data set does not distinguish if the disability is physical or mental.

Victims and Aggressors of the event that leads to the injury are obtained from the Metropolitan Police reports. Following Matthews et al. (2006), we proxy the price of alcohol using the price of beer. We obtain daily data on the price of beer by taking the average price of the different brands of beer from all the major supermarkets in the United Kingdom. The alcohol data is gathered from the “my supermarket website” on each day during the entire year of 2016.⁶ The final sample contains 365 time periods and a total number of observations of 408,000, with 93,840 observations classified as disabled and 314,160 classified as nondisabled individuals. We have an unbalanced panel as not all individuals provide sufficient details to be included in our data sample. Our sample also includes the same individuals on multiple occasions if they have been involved in more than one violent incident over the duration of our sample period. Individuals are identified by their unique National Health Service Number.

Results

Validity of All Models

We report all the empirical findings in Tables 1 to 3. First for all estimated models, we observe that the fixed and time effects of equation (1) are significant, suggesting that the individual and time-specific shocks differ significantly across the individuals in our sample, justifying the use of the panel. This is because the fixed and time effects dummies are significant across all individuals in our sample due to the fact that the p value is zero for the fixed and time effects of all models suggesting that individuals are different to each other and over time. To deal with this cross sectional and time effect, we need to estimate a model in panel format. This is why we estimate equation (1) in a panel framework.

Table 1.

The Empirical Relationship Between Disabled Individuals and Violence.

Variable	Coefficient
Constant	14.28 (0.00) *
Price of beer	−2.01 (0.00) *
Disability dummy variable	0.28 (0.00) *
$α_{i}$	(0.00)*
$b_{t}$	(0.00)*
SE	0.77
AR(1)	(0.44)
NORM(2)	(0.57)
Difference-Sargan	(0.40)
Hausman test	55.66
R ²	.25
Observations	408,000

Note. AR = autoregression; NORM = normality.

denotes significance at the 99% level.

Table 2.

The Empirical Relationship Between Disability and Violence, Where Disabled Individuals Are Aggressors of Violence.

Variable	Coefficient
Constant	16.99 (0.00)*
Price of beer	−1.23 (0.00)*
Disability dummy variable	0.13 (0.21)
$α_{i}$	(0.00)*
$b_{t}$	(0.00)*
SE	0.99
AR(1)	(0.56)
NORM(2)	(0.64)
Difference-Sargan	(0.52)
Hausman test	48.21
R ²	.20
Observations	408,000

Note. AR = autoregression; NORM = normality.

denotes significance at the 99% level.

Table 3.

The Empirical Relationship Between Disability and Violence, Where Disabled Individuals Are Victims of Violence.

Variable	Coefficient
Constant	10.23 (0.00)*
Price of beer	−1.68 (0.00)*
Disability dummy variable	0.44 (0.00)*
$α_{i}$	(0.00)*
$b_{t}$	(0.00)*
SE	0.62
AR(1)	(0.32)
NORM(2)	(0.39)
Difference-Sargan	(0.29)
Hausman test	42.66
R ²	.37
Observations	408,000

Note. AR = autoregression; NORM = normality.

denotes significance at the 99% level.

In addition, all estimated models pass the diagnostic tests. A test for first-order serial correlation is insignificant, which suggests that the panels do not suffer from serial correlation. This is because the p value of the first-order autoregression (AR(1)) test is above 0.05 for all models suggesting that we cannot reject the null hypothesis of no serial correlation. It is important that the models do not suffer from serial correlation, because then we can use the p values of conventional t statistics to determine the significance of the explanatory variables in equation (1). The Jarque–Bera normality test indicates that the residuals of the models are normally distributed, implying that the empirical estimates obtained are not due to any outliers in the data. This is because the p value is greater than 0.05 for all estimated models, suggesting that we cannot reject the null hypothesis of normality. It is important that we have normality of the error terms of the panel estimations, as this indicates that our results are genuine and not driven by a few extreme observations.

The Sargan tests confirm the validity of the instruments in all GMM system models. It is vital that we have valid instruments to obtain robust econometric estimates for the GMM system panel estimations. This is because the Difference-Sargan p values are above 0.05 in all models, implying that we cannot reject the null hypothesis of valid instruments. Finally, for all models the control variable denoted by the price of alcohol is negative and significant. This reaffirms the results of the previous literature (Matthews et al., 2006) that the lower the price of alcohol the greater the likelihood of violence and vice versa. The significance of the price of alcohol also justifies its use as a control variable in equation (1). The price of alcohol is not relevant to determining disability but is an important control variable for explaining violence. We include it as a control variable to avoid the econometric problems of omitted variable bias.

Disability on the Violence of Individuals

Table 1 looks at the impact of disability on the violence of individuals. From Table 1, the most prominent result is the positive and significant relationship between disabled individuals and the rate of violence. This is because the disability dummy variable has a p value of zero, which rejects the null hypothesis that there is no relationship between disability and violence. This implies that disabled individuals have a greater probability of being involved in an activity that causes violence.

Violence and Disability for Aggressors

Tables 2 displays the empirical association between violence and disabled individuals classified as aggressors. In Table 2, we reveal that there is no statistical association between disabled individuals and the rate of violence, when the disabled individuals cause the violence (the aggressors). We derive this conclusion because the dummy variable representing the aggressive disabled individuals in our sample is not statistically significant. This is because it has a p value of 0.21 which is above 0.05, suggesting that we cannot reject the null hypothesis of no statistical significance.

Violence and Disability for Victims

Table 3 shows the empirical relationship between violence and disabled individuals who are regarded as victims. In Table 3, we report that disabled individuals have a positive effect on the rate of violence (p value of zero), when disabled individuals are the victims of violence. We also show that the magnitude of the dummy variable coefficient is significantly larger when we only consider disabled individuals who are the victims of crime. Our results provide strong support that disabled individuals are victims of violence.

We report all the empirical findings in Tables 1 to 3. Table 1 looks at the impact of disability on the violence of individuals. Tables 2 and 3 decompose the violence data of disabled individuals into aggressors and victims, respectively.

When we decompose the disabled individuals into aggressors and victims of crime, we find some very striking results. In Table 2, we reveal that there is no statistical association between disabled individuals and the rate of violence, when the disabled individuals cause the violence (the aggressors). We derive this conclusion because the dummy variable representing the aggressive disabled individuals in our sample is not statistically significant. This is because it has a p value of 0.21 which is above 0.05, suggesting that we cannot reject the null hypothesis of no statistical significance.

In Table 3, we report that disabled individuals have a positive effect on the rate of violence (p value of zero), when disabled individuals are the victims of violence. We also show that the magnitude of the dummy variable coefficient is significantly larger when we only consider disabled individuals who are the victims of crime. Our results provide strong support that disabled individuals are victims of violence.

Discussion

The study provides strong evidence confirming the magnitude of victimization of disabled people using a large panel data set which deals with the econometric issues of endogeneity and joint determination. At the same time, our analysis demonstrates that there is no statistical association between disabled individuals and the rate of violence, when the disabled individuals cause the violence (the aggressors).

The results of the study highlight the necessity to focus on public health and other policy implications of violence using a “disability perspective” in the analysis while acknowledging the importance of disability equality in violence prevention work. Bearing in mind the oppression and violence couplet experienced by individuals with disabilities, the aim should be to break this vicious cycle of systemic inequalities and discriminatory regimes that relegate this group of individuals to the fridges of society and enhances their vulnerability to violence (Goodley & Runswick-Cole, 2011; Hollomotz, 2012).

To this end, disability-related and public-health policy formulation and implementation should place a stronger emphasis on antidiscrimination work and violence prevention by taking into consideration intricately complex interplays of dynamics that give rise to violent behaviors against disabled individuals. These violent behaviors have a genealogy and are imbued by ideological presuppositions and prejudices that need to be addressed and mitigated (Bassard, Montminy, Bergeron, Sosa-Sanchez, 2015; Goodley & Runswick-Cole, 2011). For instance, measuring disability discrimination should be at the epicenter of social analyses aimed at understanding and explaining links between violence and disability (Cea D’Ancona, 2017), while discussing implications for policy formulation and implementation.

Intersectionality-based violence prevention policies and practices are concerned with understanding the sociocultural and systemic antecedents of violence against individuals with disabilities. This process involves an informed understanding of the highly political nature of disability experience that needs intersectoral and sociopolitical interventions to eradicate the multiple forms of discrimination and oppression which have been endemic aspects of disabled people’s lives. Even though not explicitly adopting an intersectional perspective on disability violence prevention, Goodley and Runswick-Cole (2011), draw upon Zizek’s work to understand “the contours of the background” that breed violence. These contours can be traced to “the role of social relationships, institutions and culture in the constitution of violence” (p. 602). Hence, breaking the link between violence and disability necessitates challenging hierarchical power relations, oppressive regimes, and structural inequalities that are reciprocally related and converge to produce disability-related violence. These dynamics along with their spatial and temporal manifestations coalesce and increase the vulnerability of disabled people to violence.

The implementation of an intersectionality-based violence prevention program focuses on developing and implementing public-health and related policies aimed at dismantling the vicious circle of polyvictimization that is experienced by disabled individuals (Bassard et al., 2015; Busche, Scambor, & Stuve, 2012). Interdisciplinary policy interventions from health, counseling, nursing, social services, child care should be supplemented by policy interventions at the macro level (the wider social context and global socioeconomic developments; Raffo et al., 2009). Poverty, for example, can be a contributory factor to the experience of disability-related violence, hence governments should focus resources on increasing disabled people’s economic and social power, their opportunities to access education, and paid employment (Emerson & Roulstone, 2014; van der Heijden et al., 2019).

One limitation of our study relates to our sample of 26 randomly selected accident and emergency departments which represent around 33% of total accident and emergency departments in London. We were only able to obtain sufficient data from these 26 hospitals on violence and disability. This limits the study to some extent, but on the other hand, we still have a substantially larger data set than previous empirical studies in this subject area. Another limitation of our study relates to the fact that disability is self-reported. This approach has been questioned due to its biased nature that “imposes serious limitations regarding the accuracy and interpretation of the data for national estimates” (Marge, 2011, p. 154). Notwithstanding the limitations of the information obtained by self-reports, it has also been empirically documented that even though women with disabilities are largely reluctant to report violence to the police, physicians in accident and emergency departments are “the first professional from whom an abused person seeks help” (Barnet & Adler, 2009 cited in Marge, 2011, p. 154).

A further limitation of our study lies in the complexity and the reciprocal effects of variables related to victims, perpetrators, and contextual factors (Khalifeh et al., 2013). For instance, our study did not take into consideration control variables such as neighborhood crime rates or poverty. Emerson and Roulstone’s (2014) study reports on the fact that affluent people with disabilities were not more likely to experience interpersonal violence compared with their nondisabled peers, while disability-related risk factors linked to violence increased when poverty levels increased.

Moreover, our study does not report on the type of disability as respondents were not asked to specify their type of disability and other comorbid conditions they might have. The disabled/nondisabled binary constitutes a reductionist form of self-identification that ignores the complex, heterogenous, and idiosyncratic nature of the “lived” experience of disability (e.g., Thomas, 1999). This also relates to the dichotomous approach adopted for self-identification; individuals were asked to indicate—by ticking a box—whether they were disabled or not. Arguably, this approach can significantly influence the percentage of people who identify as “disabled.” Some individuals with disabilities might not consider themselves as being “disabled,” largely because their “impairment” does not constitute a central aspect of their lives and human identities (Watson, 2002). Hence, these individuals might be reluctant to accept a “disabled identity,” as they recognize their strengths and disavow arbitrary constructions of “ontological normalcy” (Annamma, Connor, & Ferri, 2013).

More empirical studies are needed to provide a more rigorous and nuanced analysis of the extent to which individuals with disabilities might experience violence not only on the grounds of their disabilities, but also on the type of their disabilities. Individuals with disabilities do not form a homogeneous group of people. The differing nature of disabilities necessitates transcending the disabled/nondisabled binary to enhance understanding of the ways in which different types of disability evoke varied responses to disability experience, as well as different degrees of privilege/underprivilege (Goodley, 2017) that mitigate/enhance the risk of victimization linked to interpersonal violence.

Simultaneously, more empirical evidence is needed to gauge the extent to which individuals with disabilities might experience violence on the grounds of the intersection of their disability status with other sources of social disadvantage linked to gender, race, ethnicity, age, sexuality, and other markers of difference (Crenshaw, 1991). Despite the availability of these studies (e.g., Casteel, Martin, Smith, Gurka, & Kupper, 2008; Emerson & Roulstone, 2014; Marge, 2011; Smith, 2008), their methodological rigor and sample population need to be strengthened so as to enhance understanding of the ways in which an intersectional analytical lens can inform violence prevention policy constitution and dissemination.

As a follow up of our study, the next step will be the examination of the extent to which differences in ethnicity and age have an impact on disability-related interpersonal violence. For example, it would be interesting to explore the interstices of disability and immigration and the ways in which disabled migrants experience interpersonal violence in their host countries. Notwithstanding the disciplinary ghettoization of literature on disability and immigration (e.g., El-Lahib, 2017; Pisani & Grech, 2015), there has been some recent, albeit limited, body of research exploring the difficulties that disabled migrants experience in their receiving countries, thereby acknowledging the cumulative and intersecting effects of racism and disablism (e.g., Hanes, 2009; Yeo, 2017). In this respect, we can hypothesize that due to their dual minority social statuses disabled migrants are more likely to experience interpersonal violence in the context of diasporas, an issue that merits empirical investigation.

Conclusion

In this article, we empirically examined the extent to which disabled individuals experience violence due to victimization. High frequency daily data based on 408,000 observations in London were collated directly from the accident and emergency units in hospitals. The GMM system panel estimator ensures that our empirical estimates are robust to endogeneity and joint determination, unlike previous empirical research in this area.

Data analysis provides strong evidence confirming the victimization of people with disabilities and the necessity to focus on disability equality in violence prevention work. Our results have implications for the monitoring process of implementing the UNCRPD (see Article 35), which, inter alia, stipulates the need to protect people with disabilities “from all forms of exploitation, violence and abuse” (Article 16). Disability-related data on people injured by violence can provide objective and evidence-based information on the extent to which the rights-based orientations of the Convention are acted upon by State Parties that signed and ratified the Convention (United Nations, 2008).

The biannual progress reports, which according to Article 35 are expected to be submitted to the Committee by each State Party documenting the measures taken to give effect to their obligations under the present Convention, should provide evidence-based information on the ways in which State Parties have facilitated the formulation and implementation of public health and other related policies aimed at challenging the multitude and cumulative forms of historical, ideological, and discursive systems of social domination and inequality that contribute to the reproduction and sometimes social legitimization and normalization of violence against people with disabilities (Bassard et al., 2015; Busche et al., 2012). Only in this way can the rhetorical proclamations of the Convention become enacted practices aimed at addressing the mutually reinforcing effects of the symbolic and physical violence experienced by individuals with disabilities.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

ORCID iD

Anastasia Liasidou

Author Biographies

Anastasia Liasidou is senior lecturer in education and social science (inclusion and disability studies) at St Mary’s University, London. Her research interests are focused on critical disability studies, inclusive/special education policy and practice, disability rights, cross-cultural and cross-disciplinary research on inclusion and social justice, critical discourse analysis. She has published more than 25 peer-reviewed articles, chapters, and books in disability studies, while being involved in a number of research and consultancy projects. Since 2014, she has published in a wide range of journals including Social Science & Medicine, Critical Studies in Education, Discourse, Emotional & Behavioural Difficulties, Disability & Society.

Andros Gregoriou is professor of finance and head of the Economics and Finance Division of the Business School at the University of Brighton. A leading authority in finance, he has published a significant number of research papers in internationally recognized journals. He has obtained more than £5 million in research funding throughout his academic career from the Economic and Social Research Council (ESRC), Leverhulme, British Academy, and Nuffield. He has also received funding for consultancy from the Bank of England, the London Stock Exchange, the CFA, and Low Carbon Futures. His research interests focus on stock market liquidity and financial econometrics.

References

Annamma

S. A.

Connor

Ferri

(2013). Dis/ability critical race studies (DisCrit): Theorizing at the intersections of race and dis/ability. Race Ethnicity and Education, 16, 1-31.

Bassard

Montminy

Bergeron

A. S.

Sosa-Sanchez

I. A.

(2015). Application of intersectional analysis to data on domestic violence against aboriginal women living in remote communities in the province of quebec. Aboriginal Policy Studies, 4(1), 3-23. Retrieved from https://journals.library.ualberta.ca/aps/index.php/aps/article/view/20894/pdf_31

Biørn

Klette

T. J.

(1999). The labour input response to permanent changes in output: An errors-in-variables analysis based on panel data. The Scandinavian Journal of Economics, 101, 379-404.

Blundell

Bond

(1998). Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics, 87, 115-143.

Busche

Scambor

Stuve

(2012). An intersectional perspective in social work and education. ERIS Web Journal, 1, 2-14.

Campbell

F. A. K.

(2009). Contours of ableism: The Production of disability and abledness. Basingstoke, UK: Palgrave MacMillan.

Casteel

Martin

S. L.

Smith

J. B.

Gurka

K. K.

Kupper

L. L.

(2008). National study of physical and sexual assault among women with disabilities. Injury prevention, 14, 87-90.

Cea D’Ancona

M. A

. (2017). Measuring multiple discrimination through a survey-based methodology. Social Science Research, 66, 239-251.

Chatzitheochari

Parsons

Platt

(2016). Doubly disadvantaged? Bullying experiences among disabled children and young people in England. Sociology, 50, 695-713.

10.

Clark

(2012). Perseverance, determination and resistance: An indigenous intersectional-based policy analysis of violence in the lives of indigenous girls. In Hankivsky

(Ed.), An intersectionality-based policy analysis framework (pp. 133-160). Vancouver, British Columbia, Canada: Institute for Intersectionality Research and Policy, Simon Fraser University.

11.

Cornwell

Schmidt

Wyhowski

(1992). Simultaneous equations and panel data. Journal of Econometrics, 51, 151-181.

12.

Crenshaw

(1991). Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review, 43, 1241-1299.

13.

El-Lahib

(2017). Theoretical dimensions for interrogating the intersection of disability, immigration and social work. International Social Work, 60, 640-653.

14.

Emerson

Roulstone

(2014). Developing an evidence base for violent and disablist hate crime in Britain: Findings from the life opportunities survey. Journal of Interpersonal Violence, 29, 3086-3104.

15.

Goodley

(2017). Disability studies: An interdisciplinary introduction. London, England: Sage.

16.

Goodley

Runswick-Cole

(2011). The violence of disablism. Sociology of Health & Illness, 33, 602-617. doi:10.1111/j.1467-9566.2010.01302

17.

Hanes

(2009). None is still too many: An historical exploration of Canadian immigration legislation as it pertains to people with disabilities. Developmental Disabilities Bulletin, 3 7, 91-126.

18.

Hankivsky

Grace

Hunting

Ferlatte

Clark

Fridkin

. . . Laviolette

(2012). An intersectionality-based policy analysis. In Hankivsky

(Ed.), An intersectionality-based policy analysis framework (pp. 33-45). Vancouver, British Columbia, Canada: Institute for Intersectionality Research and Policy, Simon Fraser University.

19.

Harrell

(2016). Crime against persons with disabilities, 2009-2014—Statistical tables. Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice.

20.

Hausman

J. A.

(1978). Specification tests in econometrics. Econometrica, 46, 1251-1271.

21.

Hollomotz

(2012). Disability, oppression and violence: Towards a sociological exploration. Sociology, 47, 477-493.

22.

Hughes

Bellis

M. A.

Jones

Wood

Bates

Eckley

. . . Officer

(2012). Prevalence and risk of violence against adults with disabilities: A systematic review and meta-analysis of observational studies. The Lancet, 379, 1621-1629.

23.

Hughes

R. B.

Lund

E. M.

Gabrielli

Powers

L. E.

Curry

M. A.

(2011). Prevalence of interpersonal violence against community-living adults with disabilities: A literature review. Rehabilitation Psychology, 56, 302-319.

24.

Khalifeh

Howard

L. M.

Osborn

Moran

Johnson

(2013). Violence against people with disability in England and Wales: Findings from a national cross-sectional survey. PLoS ONE 8(2): e55952. doi:10.1371/journal.pone.0055952

25.

Krnjacki

Emerson

Llewellyn

Kavanagh

(2016). Prevalence and risk of violence against people with and without disabilities: Findings from an Australian population-based study. Australian and New Zealand Journal of Public Health, 40, 16-21.

26.

Marge

(2011). Feasibility of expanding NEISS-AIP to create a scientific database on violence and disability. Disability and Health Journal, 4, 153-164.

27.

Matthews

Shepard

Sivarajasingham

(2006). Violence-related injury and the price of beer in England and Wales. Applied Economics, 38, 661-670.

28.

McLaughlin

(2017). The medical reshaping of disabled bodies as a response to stigma and a route to normality. Medical Humanities, 43, 244-250.

29.

Mikton

Maguire

Shakespeare

(2014). A systematic review of the effectiveness of interventions to prevent and respond to violence against persons with disabilities. Journal of Interpersonal Violence, 29, 3207-3226.

30.

Mikton

Shakespeare

(2014). Introduction to special issue on violence against people with disability. Journal of Interpersonal Violence, 29, 3055-3062.

31.

National Council on Disability. (2007). Breaking the silence on crime victims with disabilities in the United States (Joint statement by the National Council on Disability, Association of University Centers on Disabilities, and the National Center for Victims of Crime). Washington, DC: Author.

32.

Oliver

(2002). Emancipatory research: A vehicle for social transformation or policy development. Paper presented at 1st Annual Disability Research Seminar, The National Disability Authority and the Centre for Disability Studies, University College, Dublin. Retrieved from http://disability-studies.leeds.ac.uk/files/library/Oliver-Mikes-paper.pdf

33.

Olofsson

Lindqvist

Danielsson

(2015). Higher risk of vioelnce exposure in men and women with physical or sensory disabilities: Results from a public health survaey. Journal of Interpersonal Violence, 30(10), 1671-1686.

34.

Pisani

Grech

(2015). Disability and forced migration: Critical intersectionalities. Disability and the Global South, 2, 421-441.

35.

Raffo

Dyson

Gunter

Hall

Jones

Kalambouka

(2009). Education and poverty: Mapping the terrain and making the links to educational policy. International Journal of Inclusive Education, 13(4), 341-358.

36.

Ralph

Capewell

Bonnett

(2016). Disability hate crime: Persecuted for difference. British Journal of Special Education, 43, 215-232.

37.

Smith

D. L.

(2008). Disability, gender and intimate partner violence: Relationships from the behavioral risk factor surveillance system. Sexuality and Disability, 26, 15-28.

38.

Thomas

(1999). Female forms. Buckingham, UK: Open University Press.

39.

United Nations. (2008). Convention on the rights of persons with disabilities. New York, NY: Author.

40.

van der Heijden

Abrahams

Harries

. (2019). Additional layers of violence: The intersections of gender and disability in the violence experiences of women with physical disabilities in South Africa. Journal of Interpersonal Violence, 34, 826-847.

41.

Watson

(2002). Well, I know this is going to sound very strange to you, but I don’t see myself as a disabled person: Identity and disability. Disability & Society, 17, 509-527.

42.

World Health Organization (WHO). (2009). Promoting gender equality to prevent violence against women. World Health Organization. Retrieved from http://www.who.int/iris/handle/10665/44098

43.

World Health Organization (WHO). (2014). Global status report on violence prevention 2014. Geneva, Switzerland: Author. Retrieved from http://www.who.int/violence_injury_prevention/violence/status_report/2014/en/

44.

Yeo

(2017). The deprivation experienced by disabled asylum seekers in the United Kingdom: Symptoms, causes, and possible solutions. Disability & Society, 32, 657-677. doi:10.1080/09687599.2017.1320268