Respiratory Syncytial Virus Infection in Homeless Populations, Washington, USA

Jim Boonyaratanakornkit; Seda Ekici; Amalia Magaret; Kathryn Gustafson; Emily Scott; Micaela Haglund; Jane Kuypers; Ronald Pergamit; John Lynch; Helen Y. Chu

Disclosures

Emerging Infectious Diseases. 2019;25(7):1408-1411. 

In This Article

The Study

We conducted a retrospective case–control study of adults hospitalized with RSV and influenza at Harborview Medical Center (Seattle, WA, USA) during July 2012–June 2017. This center is an academic tertiary medical center that functions as the safety-net hospital for the Seattle metropolitan area. We identified patients on the basis of laboratory records of specimens containing influenza A/B virus and RSV by using a rapid PCR assay (Focus Diagnostics, https://www.focusdx.com) or the Xpert Xpress Flu/RSV test (Cepheid, http://www.cepheid.com) (Appendix).

A total of 865 patients were hospitalized with RSV infection (n = 157) or influenza A/B (n = 708) during July 2012–June 2017 (Table 1; Figure 1). We showed by multivariable analysis of risk factors for hospitalization with RSV infection versus influenza that older age, homelessness, having chronic obstructive pulmonary disease (COPD) or asthma, and drug use were associated with an increased odds ratio (OR) for RSV hospitalization compared with influenza (Table 1; Figure 2). Drug use showed a correlation with homelessness (OR 5.18, 95% CI 3.17–8.46).

Figure 1.

Detection of influenza and RSV in adults hospitalized at Harborview Medical Center, Seattle, WA, USA, July 2012–June 2017. White bars below the x-axis indicate RSV seasons; asterisks indicate weeks when cases of RSV infection peaked, on the basis of Centers for Disease Control and Prevention surveillance data in region 10 (Alaska, Idaho, Washington, and Oregon) (7,8) during 2012–2017. RSV, respiratory syncytial virus.

Figure 2.

Sociodemographic characteristics of patients hospitalized with RSV infection or influenza across 5 seasons, 2012–2017, Washington, USA. A) Age; B) homelessness; C) drug use; and D) COPD or asthma. Size of each circle indicates number of patients for that data point: small circles indicate <50 patients, medium circles indicate 50–150 patients, and large circles indicate >150 patients. COPD, chronic obstructive pulmonary disease; RSV, respiratory syncytial virus.

Overall, a higher proportion of adults hospitalized with RSV infection were admitted to the intensive care unit (ICU), readmitted within 30 days, and received any antimicrobial drug compared with patients hospitalized with influenza (Table 2). A total of 4% (7/158) of adults given a diagnosis of RSV infection died during hospitalization, compared with 3% (21/712) of those with influenza. Having COPD/asthma was not correlated with antimicrobial drug use (OR 1.07, 95% CI 0.71–1.60). Only 10% (4/40) of patients with RSV infection who were readmitted within 30 days had a positive swab specimen for the same virus at the second admission.

We sought to determine whether increased hospital readmission after hospitalization for RSV infection had other potential explanatory factors. We found by multivariable analysis that having RSV infection (OR 2.40, 95% CI 1.54–3.76) and homelessness (OR 2.06, 95% CI 1.31–3.24) remained associated with an increased odds of hospital readmission. Because homelessness and RSV infection increased the odds of readmission, persons at highest risk were homeless persons with RSV infection (OR 2.4 × 2.06 = 4.95 relative to housed persons with influenza). Age (OR 1.00, 95% CI 0.99–1.02), having COPD/asthma (OR 0.88, 95% CI 0.47–1.58), and drug use (OR 1.15, 95% CI 0.62–2.13) were not correlated with readmission.

We found that 6.5% (24,452/374,672) of all patients discharged from Harborview Medical Center during 2012–2017 were homeless. In that same period, 32% (50/157) of those with RSV infection were homeless, compared with 19% (147/708) of those with influenza (p = 0.003), 3.4% (286/8,488) of patients with a urinary tract infection (p<0.001), and 2.0% (25/1,278) of patients with an ischemic stroke (p<0.001).

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....