There is little trace of the original definition of “hospital” — a refuge for travellers — in today's restructured institutions. Scarcely a seat, never mind a comforting cup of tea, is offered to stricken family members as they huddle in emergency room cubicles and wait for the next bed to become available for their ill relative. And even our modern understanding of the word — a place where one stays to receive medical, surgical or psychiatric care — is becoming obsolete. This is regrettable, but in an era defined by escalating costs, utilization reviews and inpatient lengths of stay, the lodging that hospitals provide to the ill is understood to be incidental, and dispensible. For the most part this definitional shift, moving in sync with medical progress, has been appropriate and discreet. Occasionally it becomes visible and problematic, as when the discharged patient is homeless and has nowhere to go to recuperate.
Several studies, conducted mostly in the United States, have looked at the inpatient length of stay of homeless people in comparison with domiciled patients. One study compared over 18 000 hospital admissions of homeless adults with admissions of low-income domiciled individuals and reported that the homeless patients stayed 4.1 days, or 36% longer, per admission on average than other patients, even after adjustments were made for primary diagnosis, coexisting disease and demographic variables.1 The average cost was US$2414 per additional day spent in hospital.
That the reasons for the longer lengths of stay of homeless patients are not medical was demonstrated by a recent study that evaluated an innovative solution to inpatient lengths of stay among homeless people in California: the “hoptel,” or hospital hotel.2 There were no significant differences in adjusted mean lengths of stay between domiciled patients and homeless patients discharged to the hoptel. The authors tout the establishment of hoptels as one way of reducing inpatient lengths of stay among homeless patients.
But do we really want hospitals to re-enter the hospitality trade? How beneficial is it to contain homelessness within hospital buildings by extending or, as in the case of the old Princess Margaret Hospital in Toronto, converting their walls to provide short-term shelter?3 Medicalizing homelessness in this manner deflects attention from the poverty and housing shortages that underlie it.4
For example, in 1995 the Ontario conservative government cancelled the building of 17 000 nonprofit housing units. Only 275 housing rental units were built in Toronto last year, to meet a predicted demand for 2000 units.5 The vacancy rate in Toronto is less than 1%. The excess demand has forced the price of rental units up, leaving thousands of families in jeopardy, and the threat of homelessness one paycheque away. This pattern repeats itself throughout the country.6
We need a national, coordinated strategy to provide affordable housing for Canadians. Hospitalizing homelessness is an expensive, anachronistic solution that, unlike the appreciated cup of tea, has not withstood the test of time. — CMAJ