In STEMI, homeless patients three times more likely to die in-hospital
Patients who are homeless and present with STEMI may be up to three times more likely to die in-hospital compared with those with STEMI who are not homeless, according to research presented at the virtual Canadian Cardiovascular Congress.
Compared with those with STEMI but no homelessness, patients with STEMI and homelessness were younger, were predominantly men and were more likely to report alcohol abuse, cocaine use, current smoking and a current or previous psychiatric condition.
“Our study shows a dramatically higher rate of mortality after heart attacks in people experiencing homelessness compared to nonhomeless patients,” Samantha Liauw, MD, internist at St. Michael's Hospital in Toronto, said in a press release. “More research is needed to discover the reasons for this disparity in outcomes so that the chances of survival can be improved in this vulnerable population.”
Investigators included 2,765 patients who presented at St. Michaels Hospital in Toronto with STEMI from 2008 to 2017 with the aim of characterizing initial presentation, hospital course and clinical outcomes of homeless individuals presenting with STEMI.
According to the presentation, homeless status was determined by a review of clinical records and an administrative database. Overall, 2.6% of the cohort was identified as homeless.
Researchers observed that the rate of in-hospital mortality among patients who were homeless was higher compared with patients who were not (18.7% vs. 5.6%; P < .0001).
Compared with nonhomeless patients with STEMI, those who were homeless were:
- younger (mean age, 58.2 years vs. 62.7 years; P = .004);
- mostly men (95.9% vs. 76.5%; P < .001);
- more likely to smoke (83.6% vs. 50.4%; P < .001);
- more likely to have previous or current psychiatric conditions (29% vs. 2%; P < .001);
- more likely to report alcohol abuse (24% vs. 1%; P < .001); and
- more like to report cocaine use (15% vs. 1%; P < .001).
Patients who were homeless were also more likely present with cardiogenic shock or cardiac arrest (15% vs. 4%; P < .001) and peripheral artery disease (10.94% vs. 2.48%; P < .001) compared with the nonhomeless patients.
“The elevated risk at a younger age could be related to chronic stress from being homeless, higher rates of smoking, poverty and unreliable access to healthy food,” Liauw said in the release. “Lack of trust in the medical system, poor access to health care for chronic conditions and slower receipt of emergency therapies may also have contributed.”
In addition, the homeless cohort had a lower rate of stenting compared with the nonhomeless group (80% vs. 89.7%; P = .0071).
“In our study, the time from symptom onset to undergoing a procedure to open blocked arteries was similar between the homeless and nonhomeless groups,” Liauw said in the release. “However, I suspect that the symptom start time recorded for homeless patients may be inaccurate — for example, because they were unconscious when the ambulance was called — and that there was a longer gap before therapy. This illustrates that we need new methods to study this disadvantaged part of society.”
Anique Ducharme, MD, MSc, director of the heart failure clinic at the Montreal Heart Institute, associate professor of medicine at Université de Montréal and the scientific program committee chair of the Canadian Cardiovascular Congress 2020, said in the release: “This study is important in several aspects. First, quantifying the care gap between homeless compared to the nonhomeless patients will hopefully help to improve outcomes in this fragile population, as knowledge is needed to create solutions. Bringing these facts to the cardiovascular community might be one step in the right direction. However, there is no easy answer to such a complex situation and a coordinated effort is needed to reach out to the homeless patients and convince them to present earlier in the course of their cardiovascular condition.”
Collapse
Liauw S, et al. Abstract 12. Presented at: Canadian Cardiovascular Congress; Oct. 21-24, 2020 (virtual meeting).