Recovery Related Group Homes and Disproportionate Environmental Hazard Impacts

Introduction

Environmental justice (EJ) research has illuminated the complex relationship between various and often intersecting social inequalities and environmental harms. Early EJ scholarship demonstrated the involvement of not-in-my-backyard (NIMBY) organizations, who advocated against locally unwanted land uses often successfully driving them into less politically or economically powerful neighborhoods. ^{1 , 2} Although most research focused on the correlation between environmental harm exposure and the racial, ethnic, or economic class characteristics of communities, more recent work has also shown connections between environmental injustice and gender, age, and sexuality, with a sense that the correlation between injustice and environment is and ought to be an expanding field of inquiry. ^{3 , 4 , 5 , 6} One area that has yet to be the focus of EJ consideration is the possible relationship between addiction recovery house siting and exposure to environmental harms.

Addiction recovery is a critical issue, which should be of wide interest to those concerned with social and environmental inequalities. In the United States, prescription opioid addiction alone has reached epidemic levels despite a costly and protracted war on drugs, ⁷ which has itself been demonstrated to exacerbate existing inequalities. ^{8 , 9} Individuals seeking recovery from addiction face numerous challenges, but studies show that a main predictor of success is access to “recovery capital,” which includes inpatient treatment and a sober living environment. ¹⁰ The need for a stable sober living environment has been largely answered by sober living homes (either nonprofit or for-profit) that provide varying degrees of services and staffing. ⁷

Although sober living homes provide a valuable service to the community, they are often met with local resistance. According to the National Association of Recovery Residences, individuals in recovery are protected under the Fair Housing Act, but due to fears resulting from stigma, there is still widespread housing discriminations at local and state levels. ¹¹ Before opening in a residential area, group homes can be subject to fierce opposition from NIMBY organizations and may be forced to move into neighborhoods that are less economically or politically powerful. ^{12 , 13 , 14}

In general, NIMBY groups are active at the level of local politics. ¹⁵ Community organized, and “unwilling to accept the reassurances of technical experts and government officials,” NIMBY activists tend to comprise “the same middle-class, concerned citizens who might protest a new freeway extension or adult book store.” ¹¹ Although it may seem on the surface that NIMBY and EJ groups would have the same goal of acting locally to keep polluting industries out of residential neighborhoods, the two groups are often in opposition. In fact, a key distinction between the two groups is that NIMBY organizations may view and even act as if marginalized communities such as people of color, homeless people, or people in recovery are themselves a form of pollution. ¹⁶

In this article, we consider the intersection between environmental hazard and the siting of recovery homes. We suggest that a pattern of co-occurrence between environmental hazards and recovery houses constitutes a particular form of environmental injustice. We base our research on the assertions of scholars such as who demonstrate that the discrimination and stigma faced by people in recovery contribute to a NIMBY response when the construction or development of recovery houses is proposed. Theoretically we contend that such responses are governed by a politics of disposability, and the likening of persons in recovery with contamination is similar to the practices described by Charles W. Mills in black trash theory and in Randall Amster's work on the criminalization and exclusion of homeless people who are discursively linked with disorder, disease, and pollution. ^{17 , 18 , 19} At a material level, we demonstrate that recovery house siting exposes residents to environmental harms. We then suggest possible further areas for future EJ-driven inquiry and analysis of the substance abuse recovery.

Methods

Employing an EJ paradigm and Environmental Protection Agency (EPA) data to assess disproportionate environmental impacts faced by people in recovery residences, we focus attention on Oxford Houses in Washington, DC, and Pennsylvania. We selected Oxford House because they are a large network, with an excellent track record of best practices according to the Substance Abuse and Mental Health Administration's endorsement. They also maintain a large amount of publicly available data, such as the address for each house and a yearly report of their demographics. Oxford Houses are level I recovery residences. They function by renting residential properties and are not subject to licensing or zoning. Nevertheless, they have been the subject of controversy and have participated in numerous court battles with NIMBY organizations and municipalities, even winning cases in the supreme court. ^{20 , 21} In the case of City of Edmonds v. Oxford House, the City of Edmonds, like many municipalities, tried to use zoning practices to move Oxford House into less desirable neighborhoods. Oxford House lawyers maintained that the zoning practices were a violation of both the Fair Housing Act and the Americans with Disabilities Act.

We selected the Pennsylvania study area because the state has been fairly progressive with EJ policy and has an Office of Environmental Justice. We chose Washington, DC, as a study area because Oxford House is headquartered there. Looking at recovery residences in both Washington, DC, and Pennsylvania, we assess some of their major risk factors by using the EJSCREEN data file for 2016.

We started by analyzing the percentile that each house falls into nationally with relative exposure to cancer risk, respiratory hazard index, superfund proximity, hazardous waste site proximity, and lead paint exposure. Cancer risk information is based on the national-scale assessment (NATA) of air toxic emissions for 2011 and shows the lifetime cancer risk from inhalation of air toxins. ²² The Respiratory Hazard Index is also based on NATA and shows “ratio of exposure concentration to health-based reference concentration.” ¹⁸ Superfund proximity is calculated by the “Count of RMP (potential chemical accident management plan) facilities within 5 km (or nearest one beyond 5 km), each divided by distance in kilometers,” and based on the EPA's list of RMP sites from 2015. ¹⁸ The hazardous waste site proximity is calculated by the “Count of TSDFs (hazardous waste management facilities) within 5 km (or nearest beyond 5 km), each divided by distance in kilometers” and “Calculated from EPA's Resource Conservation and Recovery Act (RCRA) Info database, retrieved 08/16/2016.” ¹⁸ Lead paint exposure is calculated as “Percent of housing units built pre-1960, as indicator of potential lead paint exposure” using information from “Census/American Community Survey (ACS) data, retrieved 2015.” ¹⁸

We acquired the addresses for the houses in Washington, DC, and Pennsylvania from the Oxford House Directory. ²³ We then cross-referenced the addresses with the census block group they are located in by using the EJ Screen tool available on the EPA's website and compiled information from every house in both Pennsylvania and Washington, DC, to analyze the health risks associated with living in recovery houses based on the data. ²⁴ Then, using a generalized linear model with a Poisson error distribution, we modeled the expected number of Oxford Houses in a block group in both Washington, DC, and Pennsylvania, as a function of environmental hazards and other demographic indicators. ²⁵

Findings

Our analysis, graphically displayed in tables 1–6, finds that the distribution for block groups with Oxford Houses located in them was skewed to a high percentile for cancer risk, lead paint exposure, and respiratory hazard index when compared with national and state levels. ²⁶ The distribution for superfund site proximity and hazardous waste site proximity showed no significant relationship.

Cancer risk, lead paint exposure, and respiratory hazard index all showed a positive relationship with the number of Oxford Houses in a block group. Superfund and hazardous waste site proximity did not show a statistically significant relationship. The model predicted that for each additional lifetime case of cancer per lifetime per million per block group, the number of expected Oxford Houses within that block group increases by 2.8%. Controlling for both minority percentage and low income percentage, that number decreased to 2.5%. A similar model, analyzing lead paint exposure based on the percentage of homes in the area built before 1960, predicted that for each 1% increase in pre-1960 homes, the expected number of Oxford Houses in the block group increases by 3.1%. Controlling for both minority percentage and low income percentage, that number was 2.3%. In another model, analyzing scale on the respiratory hazard index, it was predicted that for each additional point on the index, the expected number of Oxford Houses located in the block group increases by 126%, and 108% when controlling for both minority percentage and low-income percentage. All effects are statistically significant.

Similar numbers emerged when running the models using the data from Pennsylvania only. The models run using the data for Washington, DC, did not show a statistically significant relationship between the number of Oxford Houses and environmental impacts. This information, however, does not disprove the hypothesis. Washington, DC, is a much smaller area of research, and is more comparable to a major city than to a state. Furthermore, Washington, DC, is a relatively toxic place to live for all of its residents. During preliminary analysis of the data, block groups containing Oxford Houses were in a high percentile for all relevant environmental impacts when compared with the national levels. When compared with Washington, DC, levels, the percentile rankings were insignificant. So the fact that the Oxford Houses in Washington, DC, did not show a significant relationship between the location of the houses and environmental impacts when comparing only to other block groups in Washington, DC, may have more to do with the nature of Washington, DC, than with the nature of the relationship between Oxford Houses and disproportionate environmental harms.

Discussion

We are confident that recovery houses are likely to be disproportionately sited in environmentally hazardous communities. Within the research area covered by this study, there is a positive relationship between block groups containing Oxford Houses and an increase of cancer risk, respiratory hazards, and lead paint exposure, even when controlling for both racial minority populations and low income, which are factors that are known to be correlated with these impacts ^{27 , 28}

More research is warranted but our preliminary findings suggest that there are serious EJ and health implications for people in recovery and their dependents who also live in sober houses. In particular, there is cause for deep concern regarding the high levels of lead paint and respiratory hazard exposure, particularly in homes where children are present as both forms of exposure have been shown to reduce academic performance. ^{29 , 30} Since people in recovery and their dependents are already at risk for ill health, delinquency, and poor academic performance, compounding factors such as lead paint exposure and respiratory hazards may constitute a further hindrance to recovery and pose a specific additional intergenerational risk when children are present. Such concerns could be the basis of important further work at the intersection of EJ and recovery.

A primary limitation of this study is its scope. Although we have looked at two areas, there are Oxford Houses and other forms of recovery houses located throughout the United States. A larger comparative study would provide a clearer picture of the relationship between treatment house siting and toxic exposures. Furthermore, we entered into our study following the logic that when a NIMBY organization is successful in keeping a polluting industry out of their neighborhood, it can wind up in a neighborhood with less political strength, such as a minority community, and that when NIMBY organizations are successful in keeping unwanted populations out of their neighborhood, that population, which already has less political strength and social power, is also pushed into the same areas as environmentally harmful industries and infrastructure. We supposed that even an organization like Oxford House, Inc., which has put up a tremendous legal battle for civil rights, would be subject to this pattern. Although the findings of our study do indeed suggest that municipalities tend to locate both services for recovery communities and environmentally polluting industries and infrastructure in the same neighborhoods, our model, however, cannot prove that sequence of events. This is where an extended case study of resistance to and the ultimate siting of recovery houses would be most useful.

In addition to case studies of NIMBY resistance to recovery houses, we also believe that there may be potential for political solidarity between EJ organizations and those who advocate for populations who are discursively and materially treated as forms of pollution (including those experiencing homelessness, mental illness, and/or people struggling with addictions). As previously noted, our research suggests that in the case of recovery houses, NIMBYism could be a mechanism for exacerbating environmental injustice, and EJ advocates may do well to begin forging alliances with other groups who are also in contention with NIMBY organizers. Studies might consider whether such alliances exist. What are the challenges and strengths of such alliances? In particular, how do these alliances develop and fare across a range of differing contexts?

We hope that our preliminary research will inspire others to develop robust analyses of the relationship between substance abuse recovery facility siting and EJ. At the organizational level, it is our hope that this type of research can aid in making recovery organizations more aware of the additional risks faced by those in recovery and may support greater environmental risk mitigation and EJ awareness within the recovery community.