Hamovitch PI
Winter 2015
A research publication of the Hamovitch Center for Science in the Human Services at the USC School of Social Work
Can big data help keep children safe? USC researchers are exploring the potential of predictive risk modeling in child welfare and well-being. A call comes in to the child welfare hotline. The caller reports that a child is being maltreated. The operator has to make a decision. Are the allegations serious enough to open an investigation? Is the child in immediate danger? What services does this family need? “Child welfare workers are often dealing with a very partial, imperfect picture of conditions that may place a child at risk of harm,” said Emily Putnam-Hornstein, an assistant professor at the USC School of Social Work. “Yet critical decisions must be made about which referrals are screened in or out and whether a case is opened for services.” What if in that moment, the hotline operator had access to a model that could scan the vast landscape of data available about that particular family, weighing risk factors such as previous involvement with child protective services and a history of substance abuse while taking into account strengths such as educational achievement and access to supportive resources? Would that information help the operator decide how to proceed? Risk | continued on page 10
Newest faculty member examines social and economic consequences of mass incarceration
2
New book explores how social workers can engage in advocacy at multiple levels of society
6
MSW graduate tackles issue of suicide among sexual minority teens and young adults
8
Vol. 5, Issue 4 Winter 2015 From the Director In the 21th century, the science and profession of social work is at the critical intersection of three revolutionary developments: a newfound appreciation of experimental science, developments in biological sciences with consequences for the social sphere, and advances in computational sciences, known as big data, which is the topic of this issue’s cover article. Big data comes in many forms. For instance, China’s Baidu search engine registers more than 63 unique visitors daily and its Kadoorie Biobank provides information on more than half a million subjects. The Cochrane Collaboration’s repository includes more than 700,000 experimental impact studies. Researchers at the USC School of Social Work are organizing and analyzing big data, which is inherently messy, fluid, and in need of careful research yet holds revolutionary and innovative promise in the service of human society. With the arrival of our newest faculty member, Robynn Cox, the school adds to its excellent team of researchers specializing in crime and justice. The United States lags woefully behind other advanced societies in terms of incarceration. In 2013, 716 per 100,000 individuals were behind bars (including 1,341 and 1,155 per 100,000 in Louisiana and Mississippi, respectively), compared to rates of 607 in Russia, 147 in the United Kingdom, and 42 in Norway. In her work, Cox considers the costs of incarcerating people at such a high rate. I hope you enjoy reading about this exciting research!
Haluk Soydan, PhD Director of the Hamovitch Center
Dean Marilyn Flynn Director Haluk Soydan Editor Eric Lindberg
Front Page Photo/iStock
Hamovitch PI is a research newsletter published quarterly by the Hamovitch Center for Science in the Human Services at the USC School of Social Work. Send questions, comments, and reprint requests to elindber@usc.edu
3
Putting people behind bars has hidden costs for society on the effects of mass incarceration on black women, including employment outcomes, family relationships, and physical and mental health. She has also worked closely with the USC Leonard D. Schaeffer Center for Health Policy and Economics as a scholar affiliated with its Minority It’s a saying that has been passed down Aging Health Economics Research Center. in Robynn Cox’s family for generations. It’s While studying economics at Duke also a maxim that has guided her research University as an undergraduate and later at on the societal and economic consequences Georgia State University, where she earned a of mass incarceration, particularly in the master’s degree and doctorate in economics, black community. she became increasingly aware of the soci“When people are marked with a criminal etal effects of criminal justice policies. record or the stigma of incarceration, there Federal grants that support the hiring is a lot of judgment passed and people give of more police officers to wage the war up on those individuals,” she said. “I’m on drugs, for instance, can lead to racial really passionate about helping people, and disparities in arrest rates. Families that experience incarceration are more likely to everybody deserves a second chance.” As the newest member of the faculty at struggle with food insecurity, regardless of the USC School of Social Work, Cox plans their employment status. to focus her research on understanding hisEven well-intentioned efforts, such as torical disparities related removing the item on job to the prison system, applications asking about exploring how to help More than $80 billion is prior felony convictions, incarcerated individuals may have unintended conspent each year on the develop useful skills corrections system in the sequences. Research has and prepare to reenter shown employers that don’t United States, according use background checks society, and calculating to the Urban Institute. the external costs of crimhire fewer black men. The USC School of Social inal justice policies and “This can be tied to incarceration. perceptions in society Work’s newest faculty A fourth-generation member is interested in of criminality,” Cox said. native of Los Angeles, calculating other costs “Because they know black she grew up in the men are more likely to of mass incarceration, Crenshaw District and have a criminal record, including social, Culver City, where her it might actually lower economic, and health childhood experiences employment. It’s not that instilled a strong interest we shouldn’t have someconsequences. in issues of social justice. thing like ban-the-box legislation, but you might “Knowing the history of incarceration and slavery and policing, also have to educate these employers about including my own personal experiences that tendency to statistically discriminate.” The disproportionate effect of incarcerahaving family members affected by that, drew me to the topic and wanting to under- tion policies on certain racial and ethnic stand it more,” she said. “When you incar- groups is clear. For example, in 2013, approxcerate a particular group, pulling so many imately 37 percent of imprisoned men were men out of a particular community where black, despite this group constituting only people are already disadvantaged, there has 13 percent of the U.S. population, according to a U.S. Department of to be some sort of social consequence.” As an assistant professor of economics at Justice report. Although Spelman College, Cox conducted research white women comprised a
A background in economics will help Robynn Cox, a new assistant professor at the USC School of Social Work, explore the social and economic consequences of mass incarceration.
An open hand gives in abundance, even as it receives.
>>
Photo/Eric Lindberg
2
larger share of all incarcerated women, the imprisonment rate was twice as high among black women compared to white women. How did these policies end up landing so heavily on this segment of the population? “This is something we are currently grappling with as a nation, particularly policing in the African American community,” Cox said. She hypothesized that the nation’s legacy of discrimination and racism, coupled with
socioeconomic factors, has generated a dual criminal justice system with a skewed balance between civil liberties and enforcement powers. To illustrate the point, Cox described how people became more willing to give up certain civil liberties in the wake of the Sept. 11 attacks to ensure their safety, such as acquiescing to increased security at airports. Race can become a wedge in that tradeoff between safety and freedom, she said.
Members of the majority may be willing to have stricter laws that increase the size of the incarcerated population because they know those laws won’t be applied in full force in their community. “Because they are not bearing the full burden, they are able to operate with their civil liberties intact and maintain these strict-on-crime policies,” she said. Using code words can also influence how people feel about criminality. When asked about combating crime in general, people tend to prefer social programs. When asked about fighting “urban” crime, they lean toward incarceration and stringent law enforcement. Dehumanization of certain groups also justifies the lack of investment in social programming, Cox said. Why invest in a population that is viewed as unfixable? “These individuals are no longer seen as worthy of having the full rights of a citizen,” she said. “This is in no way, shape, or form to say we should let murderers go free. These are all symptoms of a broader issue that we need to shed light on.” As an assistant professor at the USC School of Social Work, she plans to continue exploring this broader effect on society, including an ongoing research project with the Schaeffer Center investigating connections among health, aging, and imprisonment. One particular finding suggests that people who are incarcerated are more likely to have been knocked unconscious due to head trauma. Cox is considering how such an injury might affect everyday functioning among older adults, including the ability to maintain employment and engage in interpersonal relationships. Ultimately, as she continues to delve into these complex topics surrounding mass incarceration, Cox said she will keep her family’s advice in the back of her mind. “My mother always raised us to help others and I’ve always wanted my research to inform policy with the hope that it will make a difference,” she said. “Being at the USC School of Social Work will definitely help me achieve that goal.” t
Photo/Mike Kane
Does gang life have long-term effects on Hispanic women?
As teenage girls, they were enmeshed in the dangerous world of gangs in some of San Antonio’s rougher neighborhoods. Now approximately 15 years later, what effect has that experience had on the lives of young Hispanic women? Researchers at the USC School of Social Work will explore that question in a new $3 million study funded by the National Institute on Drug Abuse. Led by Alice Cepeda, an associate professor and principal investigator of the project, the research team will examine how intimate partner violence, drug use, and other forms of trauma have affected the physical and mental health of 300 women who were affiliated with male gang members during their adolescence. “We know that trauma like partner violence has been associated with depression and anxiety, but we don’t know the extent of the consequences for women who experience chronic, long-term exposure to this victimization,” Cepeda said. She first became aware of this unique group of gang-affiliated girls as an undergraduate research assistant studying the risks of gang involvement among adolescent Hispanic boys
as part of a project led by Avelardo Valdez, then a faculty member at the University of Texas at San Antonio and now a professor with the USC School of Social Work. At that time, most research on gangs had focused on men or adolescent boys, Valdez said. The few studies on women involved in the gang lifestyle also centered on full-fledged gang members rather than affiliates. “These gangs were embedded in the community, in family-based networks,” he said. “They drew girls in those communities into their networks — girlfriends, sisters, neighbors, and schoolmates. These girls were being exposed to the same kinds of risk behaviors.” In particular, many of these girls experienced intimate partner violence, engaged in criminal or delinquent behavior, had histories of childhood trauma, and used various illicit substances. At an average age of 16, approximately 30 percent of study participants were teen mothers.
>>
In her master’s thesis, Homegirls and Chicks: HighRisk Mexican American Gang Associated Females, Cepeda noted that these young women seemed to accept their circumstances as a normal part of life in their communities. “Many of these girls are unaware of their susceptibility to these high-risk behaviors, often times accepting them as part of their social reality,” she wrote. “In many instances, these girls are simply seeking acceptance, love, respect and prestige.” Now nearly a decade and a half later, preliminary interviews with a subset of these women have indicated they are still facing challenges. Of 32 women who participated in the original study and were recently contacted, Cepeda said 64 percent reported recent intimate partner violence, 63 percent had been incarcerated, and approximately 50 percent reported being unemployed. Cepeda and her fellow researchers, including Valdez and Hortensia Amaro, Dean’s Professor of Social Work and Preventive Medicine and associate vice provost for community research initiatives, are interested in how experiences during adolescence may have contributed to negative life trajectories among these women. They will explore rates of obesity, prediabetes, sexually transmitted infections, cardiovascular and metabolic dysregulation, and other health problems. The research team also plans to examine mental health factors such as depression, posttraumatic stress disorder, anxiety, and stress. “Hopefully we will also be able to identify factors that led some of these women to lead more conventional lives,” Valdez said. “That’s the payoff of this research.” Intimate partner violence is a major focus of the study, given its prevalence in the study population. Noting that women of color tend to go underreported in studies of domestic and partner violence, Cepeda said she suspects that cultural factors may affect whether Hispanic women seek services or report victimization. Latino culture tends to be male dominated and influenced by value of machismo, she said, and these girls were embedded in families and communities in which violence was relatively common. “It’s something that is not seen as unusual,” Cepeda said, describing how some participants in the original study responded when asked to talk about the last time they fought with their partner. “They would say, oh yeah, he got mad because he thought I was dissing him and he hit me,” she said. “He kind of slapped me or punched me, but that was it — it was nothing serious. It was obvious these young women didn’t view punching or slapping or strong holding as violence. They defined violence only if they had to go to the hospital or something more serious.” This type of trauma not only can lead to physical injuries but also has been linked to negative biological outcomes and other risk factors.
5
Amaro, who has specialized in research on the effects of relationship power and interpersonal violence, said a previous study revealed that trauma is a common thread running through the lives of women who use substances and engage in risky sexual behaviors. “Close to 100 percent of the women we saw had not only a history of interpersonal violence but also really complex trauma and multiple traumatic experiences throughout their lives,” she said. “The issue of violence was really a driving force behind their mental health disorders, including depression, anxiety, and posttraumatic stress disorder.”
“I was born and raised in those same communities. I could have easily been one of these young girls. I want to know why my trajectory took me to where I am now, compared to these young women.” Photo/Brian Goodman
4
Although it has been more than a decade since the research team has interacted with participants from the original study, Cepeda said she is confident that they will be able to reconnect with many of the women, who are now in their early 30s. During a previous follow-up study with men, they located and interviewed 122 of 160 original participants. Cepeda said some men in the original study had died, others were in federal prison and could not interact with the researchers, and a few declined to participate. She expects that it may be more difficult to convince women who have led more conventional lives to participate in the new study. “Those who have moved on may not want to be associated with the lifestyle they had as adolescents,” she said. “They may just feel that it is behind them, that they aren’t involved with gangs anymore.” However, she also expects to find many original participants in the same neighborhoods as they were 15 years ago. In a recent study, two researchers Gangs | continued on page 7
Alice Cepeda
Photo/Eric Lindberg
A new book by Bruce Jansson, who holds the Margaret W. Driscoll/ Louise M. Clevenger Professorship in Social Policy and Administration, outlines the skills social workers need to advocate for their clients at the individual, organizational, community, and societal levels.
Book makes policy advocates out of social workers Clevenger Professorship in Social Policy and Administration at the USC School of Social Work. “It opens a door to them to get into advocacy at higher levels.” In Social Welfare Policy and Advocacy: Advancing Social Justice through 8 Policy Sectors, available now from Sage Publications, Jansson extends his previous work on policy advocacy by exploring issues They might help a client apply for certain specific to topics such as gerontology, edubenefits or access needed services. Perhaps cation, criminal justice, and mental health. they assist someone with navigating the Using case studies and video vignettes, he complicated process of applying for health describes how serious issues such as violacare coverage. However, few clinicians move tions of ethical rights or poor quality of beyond that work with individuals to engage services can affect clients, families, and comat a broader level, advocating for changes in munities. In addition to analyzing the effects their organizations or society as a whole. of specific policies, the book illustrates how A new book by Bruce Jansson seeks to social workers can engage in what Jansson break that trend, offering tips and strate- terms micro, mezzo, and macro policy advogies for social workers to embrace advocacy cacy to address those problems. “People have done policy work for hunas a tenet of the profession. dreds of years, but what he has done is “It’s a tool to empower social workers matched the ethical principles of social to move beyond traditional counseling work with analysis of the core comporoles to become active in trying to nents of different policy work,” said Sam change policies,” said Jansson, who holds Mistrano, a clinical associate professor who the Margaret W. Driscoll/Louise M.
Whether they are aware of it or not, most social workers are advocates.
oversees the foundation policy curriculum at the USC School of Social Work. “Jansson has made it theoretically sound to work on policy from the social work perspective.” Although Jansson coined the term policy practice approximately three decades ago, he said social workers often view themselves as direct practitioners whose only task is to help individual clients. In addition, policy is generally seen as a dry and uninviting topic, focused largely on analysis of certain programs or understanding legislation related to social welfare. “It can feel very abstract and disconnected,” said Gretchen Heidemann Whitt, who completed her PhD at the USC School of Social Work and is currently a visiting assistant professor at Whittier College. “What does this have to do with the client who is sitting right in front of me?” Heidemann Whitt coauthored a chapter in the book on mezzo policy advocacy that outlines how social workers and other professionals in the human services can move beyond one-on-one work with
>>
Gangs | from page 5 unearthed data from the 1960s that featured interviews with individuals in San Antonio; they were able to locate many of the original participants or their descendants living in the same neighborhoods. A key finding of that study was the lack of upward mobility among Mexican Americans, Cepeda said, largely due to various social and economic factors. “People don’t really leave these communities,” she said. “I was born and raised in those same communities. I could have easily been one of these young girls. I want to know why my trajectory took me to where I am now, compared to these young women.” Valdez, also a native of San Antonio, said he has noted an increasing polarization of the Hispanic community, especially during the recent economic
7
living in Southern California. That information could be used to build interventions to prevent future generations from engaging in similar risk behaviors. For instance, Cepeda said findings might indicate that strong mother–daughter relationships are particularly critical to preventing adolescent girls from getting involved with the gang lifestyle. Perhaps interventions that focus on building resilient and supportive families Avelardo Valdez will prove most useful. “But you can’t do that if you upheaval in the United States. disenfranchised, marginalized have an environment or commu“Some people have been able population and what happens nity with no resources or opporto take advantage of changes to these people over time.” tunities,” she said, arguing for in civil rights and new opporUltimately, the researchers are larger policy changes that help tunities, but there are still large hopeful that the study will help lift these communities out of segments of the population them identify certain factors that poverty. “I hope what we do here that remain at the bottom of protect against negative out- will make some contribution to the social hierarchy,” he said. comes for not only these women reducing the health disparities “This study is a great oppor- but also others in similar circum- that exist in this population and tunity to look at this highly stances, including individuals have gone underreported.” t
clients to consider the effects of policies in organizational and community contexts. “So much of what goes on at the decision-making level of the agency affects our work with clients,” she said. “I hope this is a call to arms for these direct practitioners to become more aware and step up to the plate to become leaders.” To illustrate how a social worker might move among the various levels of advocacy, Jansson described a case study from the book involving a pregnant teenager. Although pregnant teens in public school have the legal right to decide whether or not to stay in their current school or go to a continuation school, in this particular instance the girl was not aware of that right. A social work intern at the school recognized the issue and informed the girl that she did not have to transfer to a continuation school, which might have negative consequences such as affecting her likelihood of graduating. Moving to the mezzo level, the social worker considered why the school didn’t inform the girl of her rights. “Probably because counselors don’t know the legal rights of pregnant teenagers,” Jansson said. “So she helped develop a training program for the staff.” At the macro level, the social work intern
Photo/Brian Goodman
6
“Some people have been able to take advantage of changes in civil rights and new opportunities, but there are still large segments of the population that remain at the bottom of the social hierarchy.”
delved into whether state laws are clear enough regarding the rights of pregnant teenagers and broader issues such as sex education. Jansson describes these specific manifestations of core social problems as red flag alerts and peppers the book with various examples in each of the eight sectors he examines as part of his strategy to engage readers in the world of policy. For instance, many older adults don’t receive adequate information to make an informed decision about end-of-life options such as palliative and hospice care, he said. They might not know about their legal rights regarding living wills and advance directives. “We still have roughly 16 million Americans who don’t have health insurance in the United States, even with the
“Our code of ethics mandates us to be aware of what is going on in the world and how policies affect our clients and the populations we serve.” Gretchen Heidemann Whitt
Affordable Care Act,” Jansson said. “This applies to these other sectors — people can’t afford key elements of the service so they don’t go or they are denied eligibility.” In the criminal justice sector, can groups reform the prison system, ensuring that public safety remains a priority while providing opportunities for incarcerated individuals to receive rehabilitation and skills to become contributing members of society? Heidemann Whitt, who coauthored the chapter on criminal justice, said the book offers practical approaches that will help social workers fulfill the expectation that they will advocate on behalf of their clients. “Our code of ethics mandates us to be aware of what is going on in the world and how policies affect our clients and the populations we serve,” she said. “We are also mandated to promote social justice and engage in advocacy.” “We are a reform-oriented profession going all the way back to Jane Addams,” Jansson added. “That is deep in our DNA as a social work profession and we work with many vulnerable populations that face dysfunctional policies of all kinds. I hope this book enriches social work practice and in the end I think advocacy is something we ethically have to do.” t
8
Alumnus seeks solutions to suicide risk
Photo/Eric Lindberg
A new book tracing the development of an innovative approach to addressing homelessness is now available. Housing First: Ending Homelessness, Transforming Systems, and Changing Lives is coauthored by Assistant Professor Benjamin Henwood from the USC School of Social Work in collaboration with Deborah Padgett, the lead author and professor at the NYU Silver School of Social Work, and Sam Tsemberis, who founded Pathways to Housing and developed the Housing First model. “This book demonstrates that effective policy change can and needs to occur to address many of our current day complex social problems,” Henwood said. “Los Angeles recently declared a state of emergency on homelessness. This is an apt description in many areas of our country that reminds us of the devastating effects of income inequality and signals the need for the type of transformative policy change that embodies the story of Housing First.” Described as a unique combination of evidence-based practice, the individual
the class with his colleagues trying to process and make sense of their reactions. “I developed a real fascination with how people deal with trauma,” he said. After studying how individuals respond to events such as natural disasters, terrorism, violence, and abuse, Bond spent seven years as a clinician and clinical supervisor in pediatric health care, specializing in working with young survivors of trauma. He traveled to the Indian Ocean in 2004 following the major tsunami there, helping local social workers develop protocols for postdisaster interventions with children. He eventually accepted a clinical position in San Diego, where he began supervising social work students and launched a private practice. After a few years, Bond said he realized he enjoyed teaching and supervising more than clinical work. He shifted his focus to leadership roles and spent time working
David Bond, a graduate of the USC School of Social Work and vice president of programs at The Trevor Project, is collaborating with researchers from the school to improve crisis services for youths who identify as sexual minorities.
>>
Photo/Courtesy of Oxford University Press
A possible solution to homelessness begins with Housing First
When David Bond learned that suicide is the secondleading cause of death for teenagers and young adults nationwide, he felt spurred to action. That passion only increased when he discovered that lesbian, gay, bisexual, transgender, and questioning (LGBTQ) youths are approximately four times more likely to attempt suicide than heterosexual teens and young adults. “When you hear stories about the reasons why LGBTQ young people are attempting suicide at such higher rates than their straight peers, it’s really about the way people are treated in society,” said Bond, who earned his master of social work degree at USC in 2002. Young people who identify as sexual minorities may experience hateful or marginalizing language, rejection, and even violence, he said, and they often don’t have a community or support system to rely on for help, especially if they have yet to disclose their LGBTQ status to others. Bond’s interest in suicide risk among sexual minorities brought him to The Trevor Project, a national crisis intervention organization that offers a hotline and chat and text messaging services to young LGBTQ individuals who are contemplating suicide. As vice president of programs, he recently forged a partnership with researchers at the USC School of Social Work to evaluate how The Trevor Project might be able to improve its crisis services. “David has been a big instigator to help them move toward an academic understanding of how to serve the population better,” said Assistant Professor Jeremy Goldbach, who is leading the one-year $168,000 evaluation project. “They are dealing with these silent struggles. How do we learn from them to find out what their needs are?” Born in a small town in the Midwest, Bond came to Los Angeles to pursue an undergraduate degree in international relations and psychology at USC. His interest in the mediation and negotiation aspects of international relations paired with a strong curiosity about how the mind works drew him toward a graduate degree in social work. On the day his class was scheduled to discuss posttraumatic stress disorder, the Sept. 11 attacks on the World Trade Center occurred. Bond spent much of
9
right to housing, and consumer choice, Housing First challenges the typical strategy of mandating treatment, sobriety, and other requirements before receiving independent housing. The book traces the history of homelessness in the United States, exploring how shelters and transitional housing programs became the norm during the 1980s. The linear approach to tackling issues of homelessness is then countered by the Housing First model, which provides immediate access to housing and supportive services. Developed in New York City in 1992, the model has since expanded throughout the
United States, Canada, Western Europe, and Australia. It has been credited with ending homelessness among veterans in several U.S. cities and eliminating chronic homelessness throughout Utah. In 2007, the federal government recognized Housing First as an evidence-based practice, and Canada has adopted the approach as its official policy. In addition to describing how the model operates, the book highlights research on Housing First, including findings from a study led by Padgett and Henwood in New York that involved in-depth interviews with homeless individuals who received Housing First services. The book also features first-person accounts from formerly homeless individuals regarding how their lives changed after they obtained housing and support services. Although some agencies and homelessness services providers initially greeted Housing First with skepticism, Henwood said emerging evidence of the model’s effectiveness versus standard forms of care has tempered that resistance. Research has shown that Housing First produces greater stability in housing, decreases use of drugs and alcohol, and is more cost effective than traditional approaches to homelessness. Published by Oxford University Press, the book is now available in hardcover and digital formats. t
with various youth development programs throughout the country, Project to explore issues such as intent to commit suicide, family particularly in juvenile detention, school, and health care settings. rejection, and presence of supportive peers. Goldbach said in addition to generating strategies to improve “I started to get really interested in suicide prevention,” he said. the organization’s crisis services, the project may also open the “It’s one of the leading causes of death for individuals aged 10 to 24, door to other research initiatives. and yet it is terribly underfunded.” “I think part of why they chose to go with USC is because they His interest in the topic drew him to his current role at The want to form a longer-term partnership,” he said. Trevor Project, which is considered the top crisis “The hope is with what we learn from this evaluservice provider for LGBTQ youths in the country ation, we may have some other avenues we can and engages in more than 50,000 interactions with “He is a true success pursue in collaboration.” individuals in crisis each year. story for the School Bond agreed, noting that The Trevor Project has a “One of the biggest reasons they recruited me was of Social Work. He’s wide array of data on what can be a difficult populato take an evidence-based approach to treatment a testament to the tion to access and USC has a proven capacity to conand intervention,” Bond said. types of graduates duct research on issues related to individuals in crisis. The organization had not been involved in Ultimately, both Bond and Goldbach are hopeful research or evaluation of its crisis services, he said, we develop.” that they will be able to build a successful and susand wanted to begin developing a research protainable partnership that ensures LGBTQ youths gram in collaboration with an institution that had Jeremy Goldbach in crisis receive the support they need while prorelevant expertise and resources. After The Trevor viding valuable research opportunities to improve Project solicited proposals for a comprehensive those services and other programs. evaluation, he said Goldbach’s approach emerged as the clear “That’s the balance I’m trying to strike and why it’s so helpful that winner. “I had to try to put my bias aside, but the USC proposal David is such a strong clinician,” Goldbach said. “He is a true sucwas truly the best one,” he said. As part of the evaluation, Goldbach and his team are surveying cess story for the School of Social Work. He’s a testament to the 600 randomly selected individuals who engage with The Trevor types of graduates we develop.” t
10
Photo/Eric Lindberg
An economist who has explored the use of predictive risk modeling in New Zealand, Rhema Vaithianathan (left) listens to Assistant Professor Emily Putnam-Hornstein speak during a discussion at USC about risk modeling in Los Angeles County.
Risk | from page 1 It’s an exciting new approach in the field of child safety and service delivery known as predictive risk modeling or risk analytics. In Los Angeles County, which handles approximately 200,000 allegations of child maltreatment each year, leaders are working to vet the concept. “In the child welfare system, decisions are being made throughout the life of a case,” Putnam-Hornstein said. “Should we accept this referral that came into the hotline and investigate? Are we going to open a case for services? Are we going to reunify a child who has been in foster care? Frankly, we don’t have that many tools at our disposal that would ensure any kind of standardized decision-making process.” Predictive risk modeling could take advantage of the massive amount of data being collected by various county departments to determine which children are at highest risk of negative outcomes, helping social workers filter through thousands of referrals to ensure resources are being directed to families most in need of services. The practice has been applied in
the private sector for decades, perhaps most notably in credit scoring, in which financial institutions consider various factors such as prior loan applications and credit history to assess an individual’s likelihood of making future payments on time. However, its application in the field of child welfare is more controversial due to the sensitive nature of the setting. Generating a risk score that indicates the likelihood of future maltreatment or abuse can be ethically problematic and has prompted concern about the potential for misuse. “All of this information is personal,” said Jacquelyn McCroskey, the John Milner Professor of Child Welfare. “So how do we ensure it is used to enhance the lives of children and families and communities and not to single out or target someone? We must be honest and transparent and acknowledge there are questions that are hard to answer.” Fesia Davenport, interim director of Los Angeles County’s recently formed Office of Child Protection, acknowledged those concerns during a recent convening of community leaders, scholars, and county
officials at USC to discuss the use of predictive risk modeling in the regional child welfare system. “We want to have what could be at times an uncomfortable conversation because there are implications about race and the potential for analytics to exacerbate disproportionality,” she said. “It’s the courageous discussion we need to have.” Child welfare officials in Los Angeles County, along with a majority of other counties in California and in many other jurisdictions in the United States, currently rely on a standardized set of risk assessment tools that attempt to introduce consistency and validity to how cases are handled. However, individual caseworkers are responsible for completing checklists based on statistical models, introducing the possibility of human error and unintended bias. A predictive risk model might eliminate some of those subjective biases, but could it introduce other disparities? The answer to that question depends in part on how the model is constructed. Should race and ethnicity be considered as a
>>
potential risk factor? How about living people greatly overestimate their ability in a certain low-income neighborhood? to predict future events. Even if the model proves to be relatively For instance, researchers asked doctors accurate in terms of determining the at a hospital in San Diego to determine likelihood of negative outcomes, will the whether their patients would be readresults be handled fairly? mitted after being treated and discharged. “There’s been such a long history of Rhema Vaithianathan, an economist who scores and statistics being used to foster has led research projects in New Zealand disproportionality in communities of to develop and implement risk models color,” said Jennifer Ralls, director of out- in health and child protection, said the comes and community impact at Para results were not promising. Los Niños, a local nonprofit focused on “They are no better than the toss of a promoting academic success and social coin in saying which of their patients well-being among children. “There is would come back to the hospital,” she said. already a lack of trust.” Other studies have shown that percepRalls is also a member of the Community tions of whether an event will occur are Child Welfare Coalition, a group of affected by knowledge about the situacommunity-based agencies that formed tion, said Vaithianathan, who serves as in 2012 to address concerns about the a professor of economics at Auckland county’s approach to child safety. In addi- University of Technology and senior tion to expressing a sense that community research fellow at Singapore Management members and families University. The more are not being involved in information people have, decisions such as whether “There is a real concern she said, the more likely to place a child in foster they are to predict an that data in the care, she said coalition adverse event. absence of context members are hesitant to The ultimate goal is to will outweigh human embrace predictive risk strike a balance between decision-making.” modeling. statistics and human “Will it trigger a final judgment, an issue being decision about removal Jennifer Ralls grappled with not only of a child?” she said. in Los Angeles County “There is a real concern but also in New Zealand that data in the absence of context will and Allegheny County, Penn., where govoutweigh human decision making.” ernment officials and community leaders That was a major point of discussion are also exploring the use of predictive during the recent gathering at USC, with risk modeling in child welfare. county officials and researchers empha- “We’re still a few years away from sizing that any predictive model or risk knowing if it’s the right thing to roll scoring system will not replace a social out,” Vaithianathan said. “We’re still in worker’s assessment of a particular case. research mode. I worry sometimes that Instead, the model is envisioned as a tool policy makers and practitioners, because that can help child welfare workers assess they are under so much pressure to do information from across county depart- something, just end up adopting things. ments and affiliated agencies to gain a We need to go slowly.” better understanding of the strengths and Nothing has been actively implemented risks of each family. in Los Angeles County or any other juris“I’m always the first to point out that diction, although local officials have been nobody is talking about any kind of risk testing various predictive models using modeling work that would replace clinical anonymous data from previous years to judgment,” Putnam-Hornstein said. “I create a formula that assesses the risk of don’t think we will ever see a tool that can fatal, near fatal, and critical incidents in or should replace the assessment of a well- the child welfare system. trained caseworker in the home of family.” During the recent convening at USC, However, researchers have also cau- county leaders said they are moving fortioned against relying too much on the ward with plans to release a request for ability of individual caseworkers to assess proposals from vendors to build a tool risk. Studies have repeatedly shown that Data | continued on page 12
11 Can China use big data to address social problems?
As China struggles with issues related to wealth disparity, ethnic diversity, urbanization, and internal migration, scholars are finding promise in evidence-based approaches that make use of big data. During a panel discussion at the 2015 USC Global Conference in Shanghai, experts from the USC School of Social Work and the Chinese Cochrane Center discussed how large amounts of data can be synthesized to develop effective and efficient tools that can address those societal ills. For instance, a study from 2009 indicated that more than 50 million articles had been published in more than 26,000 academic journals, with an additional 1.7 million articles produced each year. “In each area, the amount of information can be overwhelming,” said Haluk Soydan, associate dean for faculty affairs and senior fellow for global research impact at the USC School of Social Work. “That’s why systematic research reviews and meta-analyses were invented and have become more sophisticated over the past few decades.” These analytic tools help researchers combine findings from multiple studies to determine if a particular approach to a problem is effective. For example, at the Chinese Clearinghouse for Evidence Translation (www.ccetchina.org), scholars are highlighting promising interventions for aging populations that can help prevent falls, depression, and suicide. Youping Li, the founding director of the Chinese Cochrane Center, joined Soydan and Iris Chi, the Chinese-American Golden Age Association/Dr. Frances Wu Endowed Chair to discuss how China can benefit from this evidence-based approach to big data. Marilyn Flynn, dean of the USC School of Social Work and the 2U Endowed Chair in Educational Innovation and Social Work, moderated the session. Soydan said attendees responded positively to the discussion, including some Chinese students who, after participating in the session, indicated they are interested in studying at USC before returning to China to reform its health and social services system. “It’s an incremental process,” Soydan said. “They are building a lot of schools of social work but don’t really have teachers. They are looking to us to help them train teachers in evidence-based practice.” t
Hamovitch Center | 1149 South Hill Street, Suite 360 | Los Angeles, CA 90015 USC School of Social Work | Montgomery Ross Fisher Building | 669 West 34th Street | Los Angeles, CA 90089 213.821.3628 | hrc@usc.edu | usc.edu/socialwork/research
Jacquelyn McCroskey (center), the John Milner Professor of Child Welfare at the USC School of Social Work, speaks with attendees during a convening at USC to discuss the potential use of predictive risk modeling to improve child safety and well-being in Los Angeles County.
Photo/Eric Lindberg
Data | from page 11 that could be used to predict risk in current child welfare cases. “The reality is that our resources are not unlimited,” said Jennie Feria, interim executive assistant with the county’s Department of Child and Family Services. “This can help us focus in on which families are going to receive priority for some of these services.” Feria, who previously served as a supervising child social worker and later oversaw the highest-volume child welfare office in the county, with an average of 1,000 referrals a month, said results of initial tests have indicated that predictive risk modeling is a promising approach worth pursuing. “By not looking at this possibility, we would be doing our families an injustice,” she said. Officials also stressed the findings of a 2014 report by the county’s Blue Ribbon Commission on Child Protection, a group charged with reviewing recent failures in the child protection system, highlighting organizational barriers to child safety, and drafting recommendations to reform the system. In that report, the commission cited a preventive risk analytics program developed in Florida’s Hillsborough County that identified key risk factors associated with poor child welfare outcomes. According to the report, that information helped officials allocate resources more effectively to address those factors, resulting in a major reduction in child fatalities. The commission called for the implementation of a similar process in Los Angeles County, with an emphasis on ensuring that key services such as health and mental health care, substance abuse treatment, housing support, and preventive programs are being directed to families at highest risk of fatalities. “The county has a mandate to develop this work,” Putnam-Hornstein said. “The key discussion is how risk modeling is implemented and used.” One strategy to ensure that the tool is not misused is to restrict access to risk scores, perhaps only allowing the hotline operator and a supervisor to view the results of the model, she said. That would prevent caseworkers in the field from being overly concerned due to a high score or ignoring red flags during a family visit because of a low score.
“If a hotline model allows the county to identify the top 10 percent riskiest referrals, perhaps the protocol is simply that those referrals cannot be screened out without an investigation and they are assigned to a more experienced worker,” she said. “What if there was a special unit with lower caseloads and workers who had received extra training in investigations and family engagement? The score would not mean there is anything wrong with those families, simply that it is likely a more complex case that requires more time and expertise.” Along with McCroskey and other researchers at USC, Putnam-Hornstein views her role as a neutral technical resource, a sentiment echoed by community leaders such as Ralls, who said the university can serve as an academic and social justice arbiter to ensure that the community’s voice is heard throughout the process.
“I see USC as a neutral, extremely highcaliber, brilliant, and socially just institution with the best intentions to create something that can support more responsive services for kids and families,” Ralls said. Acknowledging that county officials have been increasingly open as they pursue predictive analytics, Ralls encouraged leaders to engage in ongoing meetings with the community to ensure transparency and emphasized the need for slow, thoughtful implementation. Ultimately, Putnam-Hornstein remains optimistic that predictive risk modeling will prove to be a valuable tool in child welfare, but warned against viewing it as infallible. “I think there is a lot of potential, but I don’t think we should oversell it,” she said. “Hopefully risk modeling will allow us to improve on practice as we know it today. But no tool will be perfect.” t