In this bi-weekly blog series, I explore recent research on homelessness, and what it means for the provision of services to prevent or end homelessness. Follow the whole series!


Policies tell a story. For example, workplaces requiring a doctor’s note to confirm illness tells us that there is a lack of trust around workers (recent legislation in Ontario now prevents this requirement). Undergraduate university admissions policies that are based only on grades tell the story of the absolute value of academic achievement over other experiences. 

So, what story does this policy tell: In Massachusetts, since September 2002, for families to access emergency shelter they have to either 1) experience domestic violence, 2) a natural disaster, 3) a no-fault eviction, or 4) have stayed somewhere not meant for human habitation (such as a hospital emergency department). Proof of unintentional housing loss just to access emergency shelter, a concept that seems foreign to most Canadians, I would hope.

But let’s go beyond the philosophical concerns with this policy and let’s analyze it more objectively. Knowing that families will be in need of shelter who don’t meet the first three criteria, what can they do? Well, they can bring their children to the emergency department so that the next night they can gain admission to a shelter. Sure enough, in late 2002, hospitals started to report a significant uptick in the number of families accessing shelter, particularly those who have no fixed address. 

This led Dr. Mia Kanak, Dr. Megan Sandel and colleagues to wonder about the costs to the healthcare system created by this policy; a policy grounded in distrust of families to truly need shelter. Is Massachusetts reducing shelter utilization by simply creating more costly emergency department utilization? To do so, they conducted a retrospective study of 6 years of healthcare utilization data in Boston, MA.

Results from their study are stunning: An increase in median visits per month for homeless children from 3.0 to 16.5; an increase in median length of stay from 3.1 to 8.2 hours; and over $200,000 in Medicaid costs. This is in just one emergency department in one city in the state. They estimate that what was spent to house families in emergency departments would have funded 1,594 shelter nights for families.

Here in London, Ontario, we refer to policies that have unintended consequences as “stupid rules”. The Massachusetts examples uses good research evidence to show that this requirement to demonstrate eligibility for access to shelters for families is a stupid rule, simply shifting burdens to other public systems and, ultimately, not solving the issue at hand: families left in states of homeless.