Viral Respiratory Infections in Hospitalized Children With Symptomatic Caregivers Free

A retrospective chart review was performed at a 165-bed pediatric tertiary care center with more than 6700 hospital admissions per year. More than half of the inpatient medicine and hematology-oncology units have multibed rooms, while PICU and NICU consist of single-bed rooms. Year-round, clinical staff are instructed to ask symptomatic caregivers not to stay and visitors to defer their visit. For patients under additional precautions, patients are restricted to 2 caregivers. A stricter visitor-screening policy is in place during peak viral respiratory season (typically December to March), with no child under age 14 years being allowed to visit.

All HAVRI episodes are captured by the Infection Control Program’s routine surveillance activity and entered into a safety reporting system for review and follow up by the clinical unit’s leadership team. HAVRIs reported between December 1, 2017 and July 31, 2019 in patients aged <18 years old were included in the study. A HAVRI was defined as a laboratory-confirmed acute respiratory illness, with symptoms that include rhinorrhea or congestion, cough, and fever, with onset after 72 hours of admission.⁵  Diagnostic testing for respiratory viral infections was performed in all symptomatic inpatients and consisted of a nasopharyngeal swab or deep nasal suction to detect influenza A and B, and respiratory syncytial virus (RSV). If patients were medically complex, immunocompromised, or admitted to an ICU, a multiplex respiratory panel of 16 viruses (influenza A and B, RSV, adenovirus, enterovirus, parainfluenza 1–4, metapneumovirus, coronavirus [OC43, 229E, NL63], bocavirus and rhinovirus) was performed.

The date of diagnosis was defined as the date of symptom onset and was documented in the report as the event date. Information was obtained through clinical notes and through comments on the safety report. As part of the case investigation, infection control professionals systematically review the patient chart to identify information about caregivers and visitors. Clinicians note the health status of a caregiver when they self-report or are noted to have symptoms at the time of admission and when the child experiences a change in clinical status.

Charts were reviewed and crosschecked (M.P. and N.T.), and data were anonymized and entered in REDCap. The study was granted an exemption from the hospital’s research ethics board review as a quality improvement initiative.

Forty-eight episodes of HAVRIs were reported between December 1, 2017 and July 31, 2019, with 12 events occurring in the winter of 2018 to 2019, which corresponded with the most active respiratory viral season in the province compared with other periods in the study, particularly influenza, rhinovirus and RSV.¹⁰  Two episodes were excluded as they occurred in patients aged >18 years. One case was excluded as it did not meet HAVRI case definition and 1 was excluded as chart access was restricted.

Forty-four episodes of HAVRIs in 37 patients were included in the analysis. Seven patients had 2 HAVRI episodes during their hospitalization. Most patients (n = 34, 79.5%) were admitted to an inpatient medicine unit. The remainder were admitted to the hematology-oncology unit, PICU, and NICU (6, 2, and 1, respectively). At the time of HAVRI diagnosis, patients had been admitted for a median of 28.5 days (interquartile range, 10–73.5). More than 50% of HAVRIs occurred outside of peak season (Fig 1). Rhinovirus (45.5%) was the virus most commonly identified, followed by RSV (20.5%), and influenza (13.5%).

HAVRIs occurred most frequently among patients aged <2 years and with known comorbidities (Table 1). Almost half the cases (n = 21, 47.7%) were in patients followed by the hospital’s complex care program. This program coordinates the care of patients with medical complexity or fragility, including dependence on medical technology (tracheostomy, tube feeding, supplemental oxygen requirement, etc.).

In the 72 hours before HAVRI diagnosis, 18 patients (40.9%) were in a single-bed room; of these, 6 (33.3%) were under droplet or contact precautions for another suspected (n = 2) or confirmed (n = 4) viral respiratory illness.

Ten HAVRIs (22.7%) were associated with a documented sick contact in the 7 days before diagnosis. For 9 cases (20.5%), it was with a caregiver (n = 8, 18.2%) or visitor (n = 2, 4.5%). Of note, 1 case had sick contacts with both a caregiver and a visitor. Those 9 patients had been admitted for a median of 23 days (interquartile range, 8.5–50) at the time of HAVRI diagnosis. There was no documentation of sick healthcare workers. In only 1 case, a contact with a symptomatic roommate who had been admitted with the same virus was documented in the chart of the HAVRI case. Six patients (13.6%) went home for a day pass at least once in the 7 days before their diagnosis and half of those cases were during peak season.

The most common complications associated with HAVRI episodes were increased duration of additional precautions, suspected bacterial infection with receipt of antibiotics, and increased oxygen requirements (Table 2). There were no associated deaths.

Our study shows that 1 in 5 pediatric HAVRIs were associate with an ill caregiver or visitor and underscore the importance of including family caregivers in quality improvement initiatives focused on HAVRI prevention. As the burden and outcomes in the pediatric population are significant, additional efforts should be allocated to their prevention, including caregiver screening, hand hygiene, masking for source control, and promotion of vaccination against influenza and SARS-CoV-2. As a HAVRI is known to increase the cost of an individual patient’s stay from 1.6 to 2.2 times,^3,11  financial resources invested in prevention would arguably be cost-effective.

A similar study found that 30% of HAVRIs were linked to a sick caregiver or visitor.¹²  Despite the seasonality of some viruses, the authors noted that HAVRIs could be identified year-round. In our study, HAVRIs were distributed evenly between peak and nonpeak season, suggesting that preventive measures should be applied year-round.

The study period ended before the COVID-19 pandemic; at this time, our institution has restricted nonhousehold visitors and siblings and initiated active screening among essential caregivers for COVID-19 symptoms upon hospital entry. Interestingly, we have since noted a much lower number of HAVRIs, with only 10 episodes reported between March 1, 2020 and February 28, 2021. As other factors arguably contributed to this reduction, including mandatory entry-screening for staff and universal masking for source control, this supports visitor policies and entrance screening for infectious symptoms as effective control measures.

More than one-third of HAVRI episodes occurred in patients already in a single-bed room, suggesting that a single-bed room alone is not necessarily protective and supports the implementation of other bedside strategies. Hand hygiene is universally considered an effective preventive measure for healthcare-associated infections.^13,14  However, adherence to hand hygiene by caregivers and visitors has been shown to be variable and suboptimal, ranging between 0 and 57%.^15–21  A recent study conducted in a pediatric hospital and a maternity hospital showed a low hand hygiene rate of 11% among visitors.²¹  Hospitals should engage family caregivers as partners in infection prevention and control to develop education resources to help protect children from acquiring a viral respiratory infection during the course of their admission. This is highly relevant to the current context, in which the main circulating respiratory virus in most communities, SARS-CoV-2, can be transmitted during the presymptomatic and early symptomatic periods, underlining the importance of distancing, handwashing, and masking for source control if no alternate caregivers are available at the bedside.

In our institution, during the winter season, day passes are restricted to those that are medically necessary, recognizing the risk of exposure to viruses circulating in the community. In our study, 6 patients went on a day pass in the week before their diagnosis of HAVRI and it is possible that their infection might not have been acquired in the hospital. However, there are policies to instruct the family on infection prevention behaviors as well as to screen the family and patient for symptoms and community exposures before leaving and after returning from pass.

Our findings are consistent with other studies that have shown that patients with underlying conditions, especially respiratory fragility, and of a younger age were most likely to become symptomatic due to a hospital-associated infection.^5,22  In addition to their baseline fragility, medically complex patients and infants usually require more frequent and closer contacts with healthcare workers and caregivers, which can increase their risk of acquiring HAVRIs in the absence of strict adherence to infection control measures.

Our study has several limitations. As the definition of HAVRI in our center requires a laboratory confirmation of the infection, viruses excluded from our multiplex viral respiratory panel would have been missed, potentially leading to under-diagnosis. On the other hand, a virus identified by molecular methods can reflect a remote infection and not the source of the change in clinical status. It is also possible that we missed HAVRIs in patients who developed symptoms in the 48 to 72 hours following discharge. This could have led to an over-representation of children with medical fragility as they are more likely to remain hospitalized.

Another limitation is that viral testing is not standardized in our institution, with some patients getting tested only for influenza and RSV while others for 13 additional viruses. As a consequence, there is likely an under-estimation of HAVRI’s, particularly those not associated with influenza or RSV. This ascertainment bias would impact HAVRI rates throughout the year and may contribute to the over-representation of medically complex patients observed in our study, in whom additional testing with a multiplex assay was systematically performed. Also, the surveillance definition of HAVRI is not based on virus-specific incubation periods. However, none of the HAVRIs in our study would have been classified as community-acquired based on the time from admission.²³  It is possible that we missed HAVRIs that followed 1- or 2-day incubation periods, but we expect this number to be small.

The retrospective design might have led to an underestimation of sick contacts. However, a review of sick contacts is systematically done by infection control professionals at the time of the safety report, and we likely captured most of them. We were unable to establish any causal link between caregivers or visitors and patients’ infection since adults do not receive a test to confirm their respiratory viral infection. Having a comparison group would enable us to assess the odds of developing a HAVRI when a caregiver or visitor had an illness history. Finally, the study was conducted in a single center and the findings might not be generalizable to all settings.

In conclusion, caregivers and visitors may represent a potential source of infection for pediatric patients during hospitalization. As care partners, family caregivers should receive educational resources about strategies to minimize their risks of infection transmission during their child’s hospitalization. Prospective case-control studies are required for better assessment of HAVRI risk factors.