Student Research Stories

Our PhD students are examining every aspect of social policy, from the impact of the Affordable Care Act on behavioral health disparities to the consequences of requiring pre-K and kindergarten teachers to have four-year degrees. Read more about their work: 

PhD student Joanna Taylor

The understudied consequences of school discipline

PhD student Joanna Taylor delves into the impact of suspensions on girls

“What are the consequences of school suspensions for girls aside from the school to prison pipeline?”

For doctoral student Joanna Taylor, that’s an essential, overlooked question from her years of teaching at an alternative high school in Boston.

“My whole professional background was working with youth who were fantastic people but who were really struggling in an academic setting for various reasons,” she says. “We had a lot of kids who were in foster care, who’d been through the juvenile justice system or were in various forms of probation.”

“I chose Heller because I was interested in an interdisciplinary approach,” says Taylor, who is part of the assets and inequalities concentration. “I wanted to think about the issues I was interested in in a broader context, and I've had that opportunity—to think about how education and discipline connect with health, mental health, housing, poverty and equity issues. 

School discipline for girls is an understudied area, since they receive just 30 percent of disciplinary actions nationally. But black girls have a higher suspension rate than any group except black and Native American boys.

“I came into this with a desire to look at girls cross-racially and think about suspensions in a really long-term way,” she says.

For her dissertation, Taylor is interviewing a group of black, Latino and white women in their 20s about how they remember their suspensions—which could range from a single day to multiple suspensions over a school year—and the impact that discipline had on their academic achievement, personal agency, relationships with friends and families and current situation in life. She’s also looking into the types of resources they had access to, such as church groups or adult mentors who might have helped them cope.  

She plans to compare her data to the National Longitudinal Study of Youth to better understand their outcomes.

“My main goal is to take my findings back to the communities where they are dealing with these issues on a regular basis,” she says. She hopes to present at community meetings as well as meet with school policymakers and youth advocacy organizations. “How do we create policies that create more supportive environments? When could intervention be helpful?”  

PhD student Rajan Sonik

An apple a day

PhD student Rajan Sonik shows that increases in food stamps decreased health care spending

What if doctors, faced with chronically hungry or malnourished patients, could prescribe nutritious food? Poor diets result in poor health: people who are food insecure are more likely to develop health problems. They also incur greater health care costs due to frequent trips to the doctor, pharmacy and hospital.

Second-year PhD student Rajan Sonik took this connection between food insecurity and health care costs a step further. Theoretically, he proposed, you could decrease health care costs (such as hospital admissions) by suddenly increasing in the amount of food available to poor families.

Fortunately for Sonik, in April 2009 the federal government did exactly that. As part of the stimulus package following the 2008 financial crisis, SNAP (the supplemental nutrition assistance program—formerly known as food stamps) grew dramatically. The maximum benefit level for SNAP recipients increased by 13.6 percent. 

In a study published in the American Journal of Public Health in March 2016, Sonik compared cost trends for inpatient hospital visits among Medicaid recipients in Massachusetts before and after the SNAP expansion, from 2006 through 2012. To differentiate fluctuations that are due to health care cost inflation, Sonik broke down monthly hospital cost data by number of admissions, average length of stay and average cost per day.

Sonik noted that until April 2009 health care costs were rising by about .55 percentage points per month, “which,” he says, “makes sense given the recession, which for many families meant food insecurity increased and health decreased. After the SNAP increases went into effect, health care costs were rising by only .15 percentage points per month, which is a 73 percent decline—highly statistically significant.”

These findings suggest that an investment in SNAP reduces food insecurity, resulting in improved health for beneficiaries while also providing health care cost savings.

Sonik, who advocated for children with sickle cell disease as a legal aid attorney prior to enrolling at Heller, took the study a step further. He isolated the data for patients with certain chronic illnesses: sickle cell disease, cystic fibrosis, asthma, diabetes, malnutrition, and inflammatory bowel disease.

“I compared health care costs for these patients to the regular Medicaid population, and found their health care costs were increasing rapidly prior to the SNAP increase, which makes sense since they are more vulnerable to food insecurity-based health concerns,” said Sonik. “But after their SNAP benefits increased, their health care costs actually declined.

Of his findings, Sonik says, “there are a couple of different ways to look at it. At least some people who received extra SNAP benefits experienced a positive change in their health. That implies that at least some of that money reduced food insecurity. If they had already been receiving too many SNAP dollars—which is commonly debated—or if they already had enough, we wouldn’t have seen an effect on health care costs.” In other words, Medicaid cost savings at least partially offset the increased federal SNAP spending.

Whether or not those cost savings completely neutralized the costs of those extra benefits remains unknown. “We don’t know yet whether it’s cost-effective,” says Sonik. “But at least we can say that it partially offset SNAP expansion costs, and that it’s likely even more economical for people with specific chronic illnesses or disabilities.”

A growing body of evidence suggests that social inequalities affect health disparities. “Health disparities cost a lot of money,” says Rajan. “If we’re looking at practical ways to push policymakers to make changes for families living in poverty, this type of evidence can be very powerful.”

PhD student Megan Madison

Course correction

PhD student Megan Madison explores a paradigm shift for teacher diversity and quality

“A lot of policies are made with the best intentions, “ says Megan Madison, a fourth-year Heller PhD student and former Head Start teacher, “but the way they play out on the ground can be really complicated.” This realization became difficult to ignore during her first year of teaching in Illinois, where state policy requires a greater proportion of early care and education teachers to have bachelor’s degrees. “My co-teacher had many, many more years of experience than I did, and she was teaching me how to be a good teacher every day, but I was making a living wage and she was making half of that.”

A growing body of research shows that receiving quality education in the earliest years (pre-K and kindergarten) is a critical component of a child’s healthy development. As a result, more states are focusing on professionalizing the early care and education workforce, primarily by requiring teachers to have four-year degrees.

Madison was deeply troubled by the impact this mandate would have on the early care and education workforce—disproportionately women of color—and also on the students they serve. “Are we supporting them to attain bachelor’s degrees,” she asks, “or are we pushing them out of that workforce, and replacing them with a credentialed workforce that is disproportionately white?” 

Through her PhD research, she seeks to re-frame this issue from one that pits teacher diversity against teacher quality, towards one that recognizes teacher diversity as a component of teacher quality. From a theoretical standpoint, she explores the issue’s influence on teachers and care providers, and also the students. “For young children in their first school experience, it matters that they enter a world that reflects the world they come from. They need to see themselves reflected in positions of authority,” says Madison.

Madison is conducting interviews with women of color in the early care and education world to document their experiences with professionalization policies. She’s also acquired nationally-representative data that allows her to examine the workforce both before and after implementation of these state-level bachelor’s requirements. 

As Madison progresses through the Heller PhD program, it’s become clear to her that she’s asking the right question at the right time. Concerns around K-12 teacher diversity have risen to national attention in tandem with evidence on the importance of early childhood education—a cause championed by President Obama in his call for universal pre-K at the 2013 State of the Union address. The ongoing Black Lives Matter movement has further elevated the national conversation around structural racism, inequity and injustice in the U.S.

“I think that my dissertation brings these worlds of teacher diversity, education quality and racial justice together,” says Madison. “Early care and education can be a tool for social justice, but we’re not going to get there if the system is reproducing inequities within itself. This workforce mirrors the population these programs are targeting. If we’re supporting the students and families that are the targets of these programs, how are we not supporting these teachers? We’re talking about the same people.”

When social programs reinforce social inequality

PhD student Sara Chaganti explores the effects of short-term job readiness training

In community-based career centers across the U.S., the most common strategy to help unemployed people find work is job readiness training. In these programs, participants learn to write a resume and cover letter, get interview tips and learn the basics of workplace behavior, including how to shake a hand, make eye contact and the importance of showing up on time. Over the last three decades, these programs moved away from vocational training (such as learning to operate a factory machine) and instead focus on getting people through the hiring process—and into any job—as quickly as possible. 

“Job readiness training is a really simplistic way of understanding unemployment,” says Sara Chaganti, a student in Heller’s joint PhD program in social policy and sociology. For her dissertation, Chaganti is examining these programs with an eye toward inequality. “Tons of research shows that unemployment has to do with the economy, race, gender—structural forces. But these programs are our main intervention and they don’t take any of that into account. They’re based on the assumption that unemployed people simply don’t know how to get a job. That seemed flawed to me.”

Furthermore, the workplace behavior training reinforces cultural norms that bias towards a white, middle-class presentation. “Employers say they are looking for these ‘soft skills,’ but allowing employers to dictate how an ideal worker should present means they never have to examine their own biases. They may be biased to believe that white workers are more trustworthy, or that eye contact and a firm handshake indicate reliability. Does training people to conform this way reproduce systems that marginalize people?” Chaganti asks.

After participating in these programs herself and conducting interviews, Chaganti was surprised to learn that participants really enjoy the program, saying it gave them more confidence. They also loved learning the potential reasons they weren’t getting calls back on their resumes, which helped them better understand their situation. Many participants also referenced an internal transformation, using this program as a catalyst to pivot and make positive changes in their lives and relationships. 

Chaganti also noticed moments when participants asked questions such as, ‘How can the employer get away with being a bad person?’ “Those questions were never really addressed, which seemed like a missed opportunity,” Chaganti says. “In the program, they say you have to make yourself into the kind of person the employer wants you to be. That was very troubling to me, and sad. 

As Chaganti finishes data analysis and begins writing her dissertation, she says, “One of the most important questions I learned to ask at Heller is: Are there ways that a policy, either intentionally or unintentionally, is driving inequality? I think I can safely say yes, these job readiness programs are reproducing systems that marginalize certain groups of people.”

“But I went into this research wanting to write these programs off, and I don’t feel like I can do that, because participants do get something out of it—a feeling of confidence, anyway. Mostly they get jobs paying around $12 an hour, not enough to sustain a household, but better than nothing. They’re getting out of a place of real desperation. That’s all positive.”

PhD student Tim Creedon

Did the ACA impact behavioral health treatment disparities?

PhD student Tim Creedon studies effect of insurance expansion on access to mental health and substance abuse treatment

In 2014, two major aspects of the Affordable Care Act (ACA) took effect: Medicaid expansion in 32 states, and the opening of state-level insurance marketplaces for individuals and families. The groups who stood to benefit most from these changes—the Medicaid-eligible population and those without employer-sponsored insurance—are disproportionately comprised of racial and ethnic minorities. Therefore, this expansion in health insurance could eventually both increase access to healthcare and narrow existing health disparities.

Heller School PhD student Tim Creedon and his co-author Benjamin Lê Cook set out to determine whether this is beginning to happen. In particular, they examined mental health and substance abuse treatment rates. Creedon and Lê Cook used 2014 data from the National Survey on Drug Use and Health and published their findings in the June issue of Health Affairs.

Their first question was: Did insurance coverage increase for people who need mental health and substance abuse treatment? “We found that yes, it did,” says Creedon. “Insurance coverage increased significantly for people with substance use disorders or mental health concerns—but disparities did not change.”

The same proved true for access to mental healthcare. “Mental health treatment rates increased a little bit on average for all racial and ethnic groups, which is good,” says Creedon. “But between groups, the disparities didn’t shrink or change in any big way. We saw a positive trend for Hispanics and Asians, but it wasn’t statistically significant. The disparity problem remains.” 

They then looked at substance abuse treatment rates, but saw no significant changes in 2014. One key difference between mental health and substance abuse treatment is that overall, substance abuse treatment rates are very low. “Among those people who needed mental health treatment, close to half were getting it, but less than 10 percent of those who needed substance abuse treatment received it. So we didn’t see any change in substance abuse treatment rates at all. It started much lower and it stayed much lower overall, with no changes for any racial or ethnic group.”

What this study shows is that while insurance coverage is crucial, many other factors may prevent people from accessing the care they need. Despite expansions in insurance coverage, stagnant treatment rates are a sign policy makers should focus their attention on other areas, such as treatment capacity. 

“The key takeaway for this study,” says Creedon, “is that insurance expansion is helpful, it’s necessary, but it’s not sufficient. There’s little reason to believe that insurance alone will repair disparities in behavioral healthcare access. We need to look at providers to determine if there are enough of them, if they’re located in the right communities, if they have the same backgrounds as the people that need to see them, whether they have cultural competence, and whether they accept health insurance at all. The system is huge and complicated, so we can’t expect the solutions to be simple.”

PhD student Morgan Shields

Exposing shameful quality of care at inpatient psychiatric facilities

PhD student Morgan Shields’ research seeks to uncover abusive practices and advocate for more robust surveillance systems

Over the last 50 years, inpatient psychiatric facilities have seen dramatic shifts away from institutionalization, dropping from 340 psychiatric hospital beds per 100,000 U.S. citizens to just 17. In Massachusetts alone, at least 30 state hospitals, sanitariums and state schools closed. Yet many psychiatric facilities still exist, begging the question: What is the experience like for the thousands who receive treatment in these facilities today?

Morgan Shields, a Heller School PhD, has begun to piece together data about inpatient psychiatric care from a variety of sources, but there’s very little to analyze.

“There is no national surveillance system of inpatient psychiatric facilities,” says Shields. “It’s disturbing how little we know because it inhibits our ability to improve care, appropriately regulate facilities and ensure the safety and wellbeing of consumers. 

“Outside of interviewing a large number of individuals, news articles are our best source of data on incidences of consumer harm. A media analysis can give us a broad idea of the worst cases—the ones that made it into the news.”

In her media analysis, Shields found harrowing accounts of oppression, abuse and neglect. In Jackson, Miss., individuals were admitted without screenings and kept there for years. In Pembroke Pines, Fla., youth were restrained physically and chemically for being annoying to staff, were sexually assaulted by staff and were overmedicated. In Houston, a man diagnosed with bipolar disorder was shocked with a Taser, shot and handcuffed by security guards in response to non-violent, manic behavior. The list goes on.

The closest thing to national standardized data collection for inpatient psychiatric facilities is a set of measures collected by The Joint Commission (TJC), the federal accrediting body for hospitals. When Shields analyzed this data, the performance of Veterans’ Affairs (VA) hospitals quickly stood out.

The 2014 TJC data showed that VA hospitals perform poorly when compared to for-profit, nonprofit and other government facilities. For example, at VA hospitals just 61 percent of patients receive an admission screening, well below the national average of 90 percent. The national average for appropriately justifying discharge on multiple antipsychotic medications is 53 percent, but for the VA it’s just 39 percent. Only 56 percent of VA consumers received a continuing care plan upon discharge, compared to 88 percent nationally.

“These are very basic quality measures,” says Shields. “Every facility should be doing great on these measures, and even if they are, that still would not mean that they are providing high-quality care. That’s how basic these measures are. And for hospitals that perform poorly on these measures? Yikes.” 

Top leaders from the VA headquarters in Washington, D.C. read the TJC data study and contacted Shields and her coauthor Meredith Rosenthal to discuss their findings. “They told us they’ve taken swift action, and that they’re looking into ways to improve quality and measurement,” she says.

Part of Shields’ motivation to do this research is to provoke a national conversation about quality of inpatient psychiatric care. “People are dying, people are being abused and sexually assaulted, especially youth, and I don’t hear anyone really talking about it,” says Shields. “This is actually a huge problem. I hear lots of people complaining about access to care, arguing for more beds, saying we have to get people off the streets. Fixing this is not as simple as adding some beds.”