In this episode, host Stephen Goldsmith interviews Kate Robb, who discusses her innovative, data-driven approaches to enhancing urban public health. Robb offers valuable, proven insights for city leaders looking to improve community health, safety, and trust through enhanced housing inspections and collaborative city programs.
In this episode, host Stephen Goldsmith interviews Kate Robb, discussing her groundbreaking research on how city governments can use data to improve public health. By examining the intersection of housing conditions and health outcomes, Robb shares her work in Chelsea, MA (Substandard Housing and the Risk of COVID-19), and Buffalo, NY (Tackling Persistent, Boundary-Spanning Problems Through Collaborative Innovation), demonstrating the power of innovative housing inspections, social service referrals, and collaborative city programs like Clean Sweep. This conversation provides actionable insights for city leaders on using data to create healthier, safer neighborhoods - and to become more than the sum of the parts.
Clean Sweep video mentioned in podcast: CleanSweepYIR2022.mp4 on Vimeo
Music credit: Summer-Man by Ketsa
About Data-Smart City Solutions
Data-Smart City Solutions, housed at the Bloomberg Center for Cities at Harvard University, is working to catalyze the adoption of data projects on the local government level by serving as a central resource for cities interested in this emerging field. We highlight best practices, top innovators, and promising case studies while also connecting leading industry, academic, and government officials. Our research focus is the intersection of government and data, ranging from open data and predictive analytics to civic engagement technology. We seek to promote the combination of integrated, cross-agency data with community data to better discover and preemptively address civic problems. To learn more visit us online and follow us on Twitter.
Betsy Gardner:
This is Betsy Gardner, editor at Data Smart City Solutions at the Bloomberg Center for Cities at Harvard University. And you're listening to the Data-Smart City Pod where we bring on top innovators and experts to discuss the future of cities and how to become data smart.
Stephen Goldsmith:
Welcome back. This is Stephen Goldsmith, professor of Public Policy at the Bloomberg Center at Harvard University, with another one of our podcasts, a particularly interesting one today. We have with us Kate Robb, who's a senior research associate at our center. She's a PhD from the Chan School of Public Health at Harvard. And her research focuses on how city governments can leverage data and other resources to respond to public health outcomes. So welcome Kate.
Kate Robb:
Thanks so much. It's a pleasure to be on the pod.
Stephen Goldsmith:
Thank you. Before we get to your work, just tell us a little bit about your professional background and your early research and how that was in part stimulated by COVID itself.
Kate Robb:
Sure. Well, I grew up outside of Kalamazoo, Michigan amidst a lot of forests and farmland, so not in a city at all. But I developed an interest in the environment and how the environment can shape people's health. And I studied environmental science and public health. And really for the first decade of my career, most of my work was focused on water and sanitation and hygiene in low-income countries. So I spent a lot of time in Sub-Saharan Africa and in India, working on sanitation related research in informal settlements. And eventually I became increasingly interested in environmental health in cities much closer to home.
And interested in how research can be more deeply engaged with cities in ways that can inform research and how research can lead to changes in practice in cities. And this work led me to issues of housing and neighborhoods in communities in and around Boston, where I live now. So as part of my dissertation work in public health, I started working with the city of Chelsea, Massachusetts, which is a small densely populated city just outside of Boston. And in that work I got to see really up close ways in which housing conditions were impacting the health of residents.
And I got to do this by accompanying city housing inspectors on their inspections of rental properties. The job of a housing inspector is to enforce these minimum standards for health and safety and housing. And so they go inside people's homes and they routinely encounter people in crisis and they have pretty limited tools to address the root causes of the housing code violations that they see. So they're looking for things like not having functional smoke detectors or making sure homes have sufficient ventilation or that there's no insect infestations.
But what they're seeing oftentimes are some pretty significant social or health problems. And this could be overcrowding, it could be no food in the cupboards. And more data on how housing impacts health wasn't necessarily going to help improve the situations that I was seeing, nor was the code enforcement work under the status quo that I was seeing. It was also clear that making home safer was just part of this puzzle too, that once people stepped outside of their homes, there were so many other factors that influenced their health. So densely populated, underserved areas of cities have always been vulnerable to public health risks.
It's actually one of the main reasons housing inspection came to be in the first place. So what makes cities attractive, this proximity to people and ideas also makes them vulnerable. And managing public health risks in a city is a major challenge. So using the data set that we developed on property level data in the city of Chelsea, we were able to link this with a data set from one of my collaborators at Mass General Brigham Health Systems. And by linking housing data and healthcare data, we were able to deepen our understanding of public health threats in a city and how cities can take action before, during, and after public health crises.
We used resident addresses and were able to match their housing conditions with their health records for every resident of Chelsea who was tested for COVID-19 during that first year of the pandemic. So we considered people to be living in substandard housing if they were living in a home with a history of housing code violations. And we consider people to be living in adequate housing if they had a home that had been inspected but did not have housing code violations.
And what we found was that even when accounting for other risk factors like age, sex, income, comorbidities, insurance status, that people living in substandard housing were 48% more likely to test positive for COVID-19 during the lockdown period compared to people living in adequate housing conditions. And then after the lockdown period ended, the risk for people in substandard versus adequate housing became the same. So the fact that we only observed this when people were the most exposed to their housing is significant.
The study shows how linking multidisciplinary data can be used to inform proactive solutions for cities' vulnerable residents. Cities can use this kind information to prioritize distribution of resources for future pandemics like tests and masks or even temporary housing. And as well, the results provide some further impetus for investing in longer-term solutions for safe and affordable housing.
Stephen Goldsmith:
You're only into this for five minutes, and you've got 25 different factors here. And I have all these questions for you. Let me just try to generalize for a second and then go back to your list. So I've been at this work of housing inspections for about 30 years, most of which has been unsuccessful because of the inspection process. So I think what I heard you say is you're considering the housing inspection as a indicator of other problems that could be solved as contrasted to just an inspection of the housing.
So before we get too much farther into all of the other, just talk to me a little bit about that subject. So how do we think about data that's generated in a relatively narrow vertical as an indicator of a possible solution in some other sector or some other agency?
Kate Robb:
That's a great question. So in the time I spent accompanying housing inspectors, it was really clear after a couple inspections that inspectors occupy this really unique role. They are really some of the only government officials who are inside people's homes and really get to take a look around and see what's going on as they're looking for these different kinds of housing code violations. And so while they're looking at things like wiring and ventilation, they're also seeing a lot of social and health problems.
And the only tool that they have and the only data they're recording is whether or not there's a code violation and then to issue a citation if there is one. But two things: one, oftentimes issuing a citation does little to resolve the real root cause of that problem, which can lead to a lot of other problems. And two, it's a missed opportunity to provide some other kind of service.
And so I have two different studies that I've been working on over the last several years that can, one, use the data that's generated through a code enforcement visit in new ways. And two, use that opportunity to actually intervene in a more sustainable way.
Stephen Goldsmith:
You have this terrific title "More Than the Sum of the Parts." So how would a mayor or a city leader take the data from the inspector and provide that to other agencies and organizations that will in turn then allow for more than the sum of the parts?
Kate Robb:
So one of the things that I was doing in my work in Chelsea was working hand in hand with housing inspectors to develop and implement a social service referral program within housing inspection. And this really came about through some of the things that I've already described, but that housing inspectors were finding people living in enclosed porches with no heat or running water or sanitation. And they needed other tools. When I was with them, they would find a house that was completely hoarded and had blocked egresses, which is a code violation, but it's also a behavioral health problem.
And issuing a citation for a behavioral health problem isn't going to resolve the underlying cause. And so as part of my dissertation work back in 2018, I worked really closely with housing inspectors to develop and implement a social service referral program within housing inspection. And this was a really novel program and a big shift in the way inspectional service departments usually operate. So it took a lot of time and a lot of relationship and trust building.
But now in the city of Chelsea, when inspectors encounter a problem that can't be fixed through code enforcement alone, they then are able to make a referral to a case manager who then connects the resident with a form of social assistance that they might need. So it could be for behavioral health, it could be for substance use treatment, it could be for helping to find a job, housing relocation assistance, a huge range of needs.
But what it's done for inspectors is really take this psychological burden and increase workload off their plates because they're able to refer some of the most intractable problems and focus on what they're good at, which is the inspection, but then it's also getting at the root cause of these problems and helping to prevent further crises.
Stephen Goldsmith:
So let's take that as an opportunity for you to tell us a little bit about the work you did or studied in Buffalo, New York. And over the years, maybe 10 years ago, the Innovations Program at Harvard gave an award to a Los Angeles effort that looked at code inspections in terms of when the owner of the property deserved a penalty, think slumlord, as contrasted to where the owner of the property had a behavioral problem, in your words, or a financial problem that prevented them from actually resolving the issue.
So if we think about blight remediation in Buffalo in categories such as that, tell us a little bit about the Clean Sweep project that you wrote about.
Kate Robb:
For the last five years, I've been working closely with Oswaldo Mestre, who's the director of citizen services in the city of Buffalo and advisor to the mayor. And he leads this initiative called the Clean Sweep. And the goal of the Clean Sweep Initiative is to improve quality of life for Buffalo's poorest residents. And what's really neat is that the program brings together all different departments of city hall along with community organizations into the neighborhoods.
And like you mentioned, at the beginning, departments that are focused on enforcement, departments that are focused on service provision, departments that are focused on both, they're all working together at the same time and in the same place. And what they're doing is really rapid and intensive blight remediation, community outreach work, education and code and law enforcement work. And so they meet every Wednesday morning in an area of about two blocks in a different place in the city each week. And they do things like trim trees, board up vacant properties, remove literal tons of trash.
They greet residents, they listen to their concerns, that residents share information with them. Community police officers are there to talk about community safety programs to get to know residents. It's a really neat program. If we can, I would love to include a video of it in the show notes because it's something you have to really see. It makes a difference to really see it, to see this concentration of city staff all out in the neighborhood working together at the same time.
And this is in areas of the city that have historically been incredibly marginalized and disconnected from local government. So it's a big deal to see this kind of city presence.
Stephen Goldsmith:
What if you were a city leader, say in Buffalo or anywhere in an ideal program like this, what would your metrics be that you wanted to reward in terms of neighborhood responsiveness, cleanliness? How would you measure success?
Kate Robb:
So we wanted to look at the impact of this program, and we wanted to do so in a way that we could take advantage of some existing city data. So data that cities are already collecting that then could be aggregated at property level. And so we looked at two things with this. One was really how this program works from a collaborative innovation standpoint. And another was what's the impact of a program like this? So as far as the impact, what we were able to do was put together a match sample of properties going back to really about a decade ago until 2022.
And we looked at properties that had received a Clean Sweep and properties that hadn't, and we were able to match them based on a lot of their known characteristics. And for the outcomes, we looked at 311 calls and 911 calls. 311 being an indicator of a service need to the city and 911 as an indicator of reporting crime. And what we found was properties that had received a Clean Sweep in the six months following that Clean Sweep, they were more likely to report drug related crime.
And in fact, they were 42% more likely to report drug related crime. And we also found that they were 9% more likely to report blight related service needs in this period of six months after the Clean Sweep compared to statistically very similar properties that had not received a Clean Sweep.
Stephen Goldsmith:
Hold on, this sounds like a big deal to me. So that intervention in Clean Sweep, I think this is your proposition, it encouraged the community to participate with the city. The Clean Sweep you might think of as producing more trust in the neighborhood potentially.
Kate Robb:
So we actually heard a lot about this from our interviews and focus groups with residents in the city of Buffalo and with city staff in Buffalo. Through analysis of the qualitative data and through literature on reporting behavior, we were able to identify three main reasons for why we think this reporting of drug crime and blight-related service needs went up after an intervention like this. And we don't think it's because actual crime or actual blight increased after the intervention.
So public safety, the quality of public spaces really depends a lot on residents reports of crime and service needs. And cities need this information to deploy resources effectively and equitably. But government responses and treatment of residents often differs by racial groups and by economic groups. And this has a big impact on reporting of these issues and this can really deepen disparities in neighborhoods.
So what we found through talking with residents and staff was that these changes brought about during the Clean Sweep made residents more willing to collaborate with local government in maintaining public safety and public spaces. And we identified three mechanisms through which this was mainly happening. One was government demonstrating responsiveness to the concerns of the community. Second was by building trust through these conversations between city staff and residents.
And through residents increasing in self-efficacy in how they respond to seeing social disorder like drug activity and physical disorder like blight in their neighborhoods. The Clean Sweep, it's relatively simple in its concept. It's harder to execute, but it's this relatively simple environmental intervention with community engagement that can help bridge this really critical trust gap between historically underserved communities and local governments to make communities safer and healthier.
Stephen Goldsmith:
So let's just play a little imagination game here. Let's say you are standing in front of a city's 25 housing code inspectors and you are about to challenge them about how they are going to move from being housing inspectors to ambassadors for quality of life. So this is the new group of ambassadors. What are you going to tell them to do or look for when they go into a home to lead the city's efforts for a higher quality of life?
Kate Robb:
One of the important things about this is that housing inspectors for a program like this can't be the targets of a change. They really need to be the agents of leading this kind of change. They have to be convinced that looking for things beyond code violations isn't going to increase their workload, and in fact can help to reduce their workload. And that by seeing and really no longer turning a blind eye to some of these problems that they didn't have tools to deal with in the past, by actually addressing those issues head on, it can reduce some of their psychological burden.
Housing inspectors would talk to me about some of the challenges in going home at night knowing that a kid was unsafe, or knowing that this woman was no longer really safe living in her apartment. I think it's about really re-imagining that role of housing inspectors, but it has to come from the housing inspectors themselves wanting to take on that kind of role, which takes time, but is absolutely doable.
Stephen Goldsmith:
Just a couple more questions for you and then we'll let you go. One set of questions deals with what data would you use to help preemptively identify the most at-risk homes? And the second would be, once you identify those homes and the inspector is inside them, what would he or she look for that would lead to the best public health interventions?
Kate Robb:
Thanks. That's a great question. City departments are collecting all kinds of data for various purposes. There's tax data that might have the value and age and size of homes. There's data on 911 calls, 311 calls to a property if they have permits for construction, if they're behind on water bills. And these data come in a whole variety of formats. But what's nice is that they're often in a city associated with an address or a parcel, and that makes them possible to link to each other.
And so by aggregating data on properties in a city, which is really no easy feat at all, beyond the technical challenges of that, there's the behavioral adaptive challenges of getting people to share data across departments. But when you have a data set that has a lot of information about each property, you can bring it together and to start to answer questions that can inform more strategic decisions for a city. For example, in Chelsea, we use data on each residential property in the city.
And we selected a subset of properties that had been proactively inspected for housing code violations, the kind that might cause health or safety concerns for residents. And we use machine learning to train and test a model that could be used to identify or make a prediction about the likelihood of a property having a housing code violation that might impact the residents or neighborhoods health. And so we applied the model to all the residential properties in the city, about 70% of the city is rental properties.
And we found that if the city were to adopt a more data-driven approach to housing inspection, that they could identify homes with housing code violations about twice as fast as they could using their conventional practices. And conventional practices for the city of Chelsea is going really house by house or looking at the outside of a property to make a guess about what the inside conditions might be. And I should say too, that Chelsea already had a pretty best practice form of housing inspection, which is a proactive program.
In most cities, the majority of inspections are triggered by complaints. And in a model like that, really the most vulnerable residents are the least likely to file a housing complaint due to a variety of reasons, like fear of landlord retaliation and raised rents, disclosure of immigration status, or just not knowing that they can file a complaint. But proactive programs seek to overcome this, but it also treats every property as equally risky. And we know that's not the case. And so if we use data, we can, in this example, identify properties twice as quickly that have these housing code violations.
Stephen Goldsmith:
Once they're identified and you're inside them. What would be the top one or two things that a code inspector trained as a public health inspector would look for? If I'm a mayor, I have a limited budget. I want to use that budget to intervene where I can make the greatest difference. What would the inspector look for inside the home that would provide a high value return on their time?
Kate Robb:
I think this is another place where the data that housing inspectors are already collecting can make a big difference. So there's all kinds of health conditions or health risks that are associated with hazards inside the home. For example, the city of Boston has a program focused on looking at asthma triggers inside the home. And that can make a big difference in a community where there's a lot of seniors, looking for trip hazards, which are really common, and some of those can constitute housing code violations.
And so using some of the data that's available to look at what some of the risks might already be can help. And then also looking at, through partnership with health systems, with community organizations, what are some of the prevalent health conditions and what can be done at the household level to reduce those risks. I would say those are the most important things for inspectors to be looking at. And that can really vary by city or by neighborhood within a city.
Stephen Goldsmith:
I want to switch subjects just for a second. You wrote this terrific piece, article, "Tackling Persistent Boundary Spanning Problems Through Collaborative Innovation: Lessons from Clean Sweep." And I was attracted to a sentence, "Cross-boundary teams often get stuck in early stages of collaboration because they cannot agree on a shared vision sufficiently compelling to overcome the obstacles."
That caught my attention for two reasons, one because it's so insightful. Two, because with my co-author, Kate Coleman, we wrote a book called Collaborative Cities: Mapping Solutions to Wicked Problems. Presenting the theory that place-based thinking allows for the foundation for collaboration. So if you're going to bring together not just city agencies, maybe you have a county agency in terms of public health, maybe you have Meals on Wheels or a child care agency or the like. How would you organize a collaboration to take on the resolution of the problems we've been discussing?
Kate Robb:
One of the neat things that the Clean Sweep in Buffalo does is it encourages city staff to begin with a problem, not begin with what solutions they already have available, but really start with that problem. And oftentimes this means hearing directly about the problem from the person experiencing the problem. And not just having the frontline workers, like housing inspectors, hear about the problem, but also having city office staff and city leadership hearing alongside the frontline workers about that problem.
And when you have those kinds of different departments and different levels hearing about the problem together, it can really fuel this co-productive dynamic that can enable much better problem solving.
Stephen Goldsmith:
If you look back at Chelsea and Buffalo and the theories that you have espoused, how do you think about data analytics as informing the conduct of public officials? We've written a lot about how technology should allow a city to move from routines to preemption. To move from just going from house to house to house, to using the data to more focus the work that's done. So just kind of last question, what's the role of data and analytics in this predictive and preventative scheme that you have advocated?
Kate Robb:
I think data plays a huge role in this, and so much of the data to improve the ways that cities work really already exists within cities. So they're already generating that data. It's often a matter of being able to bring that data together in new ways across departments and having the capacity to examine it, to link it. And to start to think about new questions that can be asked with that data and new strategies that could be informed by that data.
There's a really important role for increasing the capacity of cities to work with the data they already have, and that also means increasing the capacity of cities to collaborate with each other and to be willing to try new things here.
Stephen Goldsmith:
Your work is exciting. We have spent a lot of time thinking about environmental conditions and their effect on public health, time spent trying to determine how predictive and preemptive actions can improve the return on investment for city activities. Your work is indeed exciting as is your writing because it provides clues to mayors and city leaders about how to improve the quality of their neighborhoods and public health. So let me just say we are delighted that you have spent time with us today.
I encourage folks to look at your writing and your articles on this subject and follow your work. This is Steve Goldsmith, professor of Practice on Urban Policy at the Bloomberg Center for Cities with Kate Robb, senior researcher at the Bloomberg Center. Thank you Kate for your terrific work and your time today.
Kate Robb:
Thanks so much.
Betsy Gardner:
If you liked this podcast, please visit us at datasmartcities.org, find us on iTunes, Spotify or wherever you get your podcasts. This podcast was hosted by Stephen Goldsmith and produced by me, Betsy Gardner. Thanks for listening.