Hospitalizations and outpatient visits for rhinovirus-associated acute respiratory illness in adults

Abstract

Background: Rhinovirus is linked to asthma exacerbations and chronic obstructive pulmonary disease exacerbations in adults. The severity and rates of rhinovirus acute respiratory illnesses (ARIs) in adults are uncertain.

Objectives: We sought to determine rhinovirus-associated ARI rates in adults presenting for care in multiple settings and identify factors associated with rhinovirus detection.

Methods: This prospective, population-based cohort enrolled Tennessee residents 18 years or older in the emergency department (ED), outpatient clinics, or hospitalized for ARI from December 2008 to May 2010. Nasal/throat swabs were collected and tested for rhinovirus and other viruses by using RT-PCR. Rates of ED visits and hospitalizations were calculated and rhinovirus-positive and rhinovirus-negative patients were compared.

Results: Among 2351 enrollees, rhinovirus was detected in 247 (11%). There were 7 rhinovirus-associated ED visits and 3 hospitalizations per 1000 adults annually. Patients with rhinovirus, compared with virus-negative ARI, were more likely to present with wheezing (odds ratio [OR], 1.7; 95% CI, 1.23-2.35; P < .001), to be a current smoker (OR, 2.31; 95% CI, 1.68-3.19) or live with a smoker (OR, 1.72; 95% CI, 1.10-2.67), have a history of chronic respiratory disease (OR, 1.61; 95% CI, 1.17-2.22), and were less likely to be hospitalized versus seen in the outpatient setting (OR, 0.58; 95% CI, 0.41-0.83).

Conclusions: Rhinovirus is associated with a substantial number of ED visits and hospitalizations for ARIs in adults. There may be modifiable factors that can reduce the likelihood of presenting with rhinovirus-associated ARIs.

Keywords: Rhinovirus; acute respiratory illness; adults; emergency department; hospitalized; smoking.

Figure 1. Seasonal prevalence

Among adults seeking care for respiratory illness or fever, rhinovirus (RV) detection differs significantly by month. A) When all rhinovirus-associated episodes are examined, virus prevalence is highest during August and September of 2009 (p < 0.001). This graph shows the probability of a rhinovirus-positive ARI. B) When probability of rhinovirus C ARI is compared to rhinovirus A, peaks of rhinovirus A are seen during August 2009 and April 2010; however, differences are not statistically significant (p = 0.16). Shaded regions represent 95% confidence intervals.

Figure 2. Hospitalization

Virus status does not account for hospitalization of subjects. A) Odds ratios and confidence intervals of hospitalized subjects compared to outpatients and emergency department visits adjusted for viral status (p = 0.009), sex (p < 0.001), smoking status (p = 0.05), public insurance (p < 0.001), cardiac disease (p < 0.001), oxygen use (p < 0.001), chronic respiratory disease (p < 0.001), chronic steroid use (p < 0.001), and B) age (p < 0.001). Older adults had an increased probability of hospitalization.

Figure 3. Wheezing

Rhinovirus (RV) status is associated with increased odds of wheezing. A) Odds ratios and confidence intervals of subjects with wheeze compared to those without wheeze are shown, adjusted for rhinovirus (p < 0.001), other virus positive (Influenza, metapneumovirus, respiratory syncytial virus; p < 0.001), sex, smoking status (p < 0.001), public insurance, cardiac disease, oxygen use, chronic respiratory disease (p < 0.001) and chronic steroid use. B) Age is also associated with wheeze (p < 0.001), the probability of wheeze increases with age until approximately 45 years old, then gradually decreases.

Figure 4. Rhinovirus-positive and ‘other virus’ positive subjects

A) Smoking status (p < 0.001) and chronic respiratory disease (p = 0.02) increase odds of being rhinovirus (RV) positive compared to being virus negative. B) Odds ratios for subjects that are ‘other virus’ positive compared to virus negative, and C) probability of being rhinovirus-positive (dashed line) or ‘other virus’ positive (solid line) with age (P = 0.003). Odds ratios (midpoint) and confidence intervals are shown. Values were adjusted for age, race, smoking status, public insurance, cardiac disease, oxygen use, chronic respiratory disease, chronic steroid use, and living with children under the age of 18.