The Oregon Child Absenteeism Due to Respiratory Disease Study (ORCHARDS): Rationale, objectives, and design

Abstract Background Influenza viruses pose significant disease burdens through seasonal outbreaks and unpredictable pandemics. Existing surveillance programs rely heavily on reporting of medically attended influenza (MAI). Continuously monitoring cause‐specific school absenteeism may identify local acceleration of seasonal influenza activity. The Oregon Child Absenteeism Due to Respiratory Disease Study (ORCHARDS; Oregon, WI) implements daily school‐based monitoring of influenza‐like illness‐specific student absenteeism (a‐ILI) in kindergarten through Grade 12 schools and assesses this approach for early detection of accelerated influenza and other respiratory pathogen transmission in schools and surrounding communities. Methods Starting in September 2014, ORCHARDS combines automated reporting of daily absenteeism within six schools and home visits to school children with acute respiratory infection (ARI). Demographic, epidemiological, and symptom data are collected along with respiratory specimens. Specimens are tested for influenza and other respiratory viruses. Household members can opt into a supplementary household transmission study. Community comparisons are possible using a pre‐existing and highly effective influenza surveillance program, based on MAI at five family medicine clinics in the same geographical area. Results Over the first 5 years, a‐ILI occurred on 6634 (0.20%) of 3,260,461 student school days. Viral pathogens were detected in 64.5% of 1728 children with ARI who received a home visit. Influenza was the most commonly detected virus, noted in 23.3% of ill students. Conclusion ORCHARDS uses a community‐based design to detect influenza trends over multiple seasons and to evaluate the utility of absenteeism for early detection of accelerated influenza and other respiratory pathogen transmission in schools and surrounding communities.

household transmission study. Community comparisons are possible using a preexisting and highly effective influenza surveillance program, based on MAI at five family medicine clinics in the same geographical area.
Results: Over the first 5 years, a-ILI occurred on 6634 (0.20%) of 3,260,461 student school days. Viral pathogens were detected in 64.5% of 1728 children with ARI who received a home visit. Influenza was the most commonly detected virus, noted in 23.3% of ill students.
Conclusion: ORCHARDS uses a community-based design to detect influenza trends over multiple seasons and to evaluate the utility of absenteeism for early detection of accelerated influenza and other respiratory pathogen transmission in schools and surrounding communities. hinges on early detection and recognition of outbreaks. [4][5][6] Existing influenza surveillance programs rely on medical facilities to report cases of influenza-like illness (ILI) and test-confirmed influenza. 7,8 Even though influenza transmission among school-aged children frequently precedes subsequent community transmission, 9,10 there have been no systematical evaluations of school-based monitoring of influenza activity for complementing routine surveillance or serving as an early-warning system for increased influenza activity in the wider community. Monitoring school absenteeism is feasible, as seasonal outbreaks occur between late fall and mid-spring while schools are in session, 11,12 and most of the 13,588 school districts 13 across the United States collect daily absenteeism data using electronic school information systems. 14 During the 2009 influenza pandemic, there was high correlation (r = 0.92) reported between hospitalized influenza cases and school absenteeism due to ILI in one jurisdiction, 15 likely enhanced by the short, intense nature of the outbreak, which amplified absenteeism related to ILI. Conversely, monitoring all-cause absenteeism was less effective due to the multifactorial nature of absenteeism. [16][17][18] The value of continuously monitoring cause-specific absenteeism over the entire school year to identify local activity acceleration of seasonal influenza is not well understood.
The goal of Oregon Child Absenteeism Due to Respiratory Disease Study (ORCHARDS) is to develop and implement a system for daily school-based monitoring of ILI-specific student absenteeism in kindergarten through grade 12 (K-12) schools and assess the system's usability for early detection of accelerated influenza and other  89.43 W). 19 The region experiences distinct temperate seasonality.

| Community
The OSD population is estimated at 20,094 and is less racially and ethnically diverse, wealthier, and more formally educated than the United States average, while ages of individual family members and mean household size are similar (Table 1). 20 The district is composed of 6 public schools enrolling 4091 students, including prekindergarten students, in 2018-2019. 21 There are 3 elementary schools (K-4: 1503 students), 1 intermediate school (5-6: 623), 1 middle school (7-8: 596), and 1 high school (9-12: 1145).

| Timeframe
Initiation of ORCHARDS data collection occurred in phases. Absenteeism data collection commenced on September 2, 2014. Data collection from student home visits commenced on January 5, 2015. The household transmission substudy started on January 6, 2016.
F I G U R E 1 Theoretical framework of ORCHARDS demonstrating the relationships between influenza in school-aged children, K-12 school absenteeism, and medically attended influenza in the community. The relatedness of the four components (C1-C4) of ORCHARDS and the three primary hypotheses (H1-H3) are provided

| Absenteeism monitoring system
We modified an existing absenteeism reporting system. Parents report unscheduled absences using an automated telephone system, providing the student's name and the reason for absence, including symptoms if the child has a cold or flu-like illness. Uniform messaging is on each school's absentee line: Please inform us if your child has any flu-like symptoms such as fever with cough, sneezing, chills, sore throat, body aches, fatigue, runny nose, and/or stuffy nose.
In the event that a student is absent without a report, OSD attendance staff make repeated efforts to contact the home or parents/ guardians before the end of the day.

| Absenteeism definition
Because of variability among schools in terms of the number of class periods for which a child can be absent and for simplicity/ generalizability of electronic data retrieval, we consider a student absent for the entire day if absent for any part of a school day.

| Types of absenteeism
All-cause or total absenteeism (a-TOT) is an absence for any reason.
Absence due to illness (a-I) is an absence due to any reported illness.
Absence due to ILI (a-ILI) is an a-I for which ILI symptoms are reported.

| a-ILI definition
We considered established definitions for ILI 22,23 and used a simplified version of the Centers for Disease Control and Prevention (CDC) standard definition. ILI for ORCHARDS is defined as presence of fever and at least one respiratory symptom (cough, sore throat, nasal congestion, or runny nose).

| Data system
OSD utilizes Infinite Campus ® , 24 a commercially available, electronic student information system (SIS), to track student attendance. This system allows attendance staff to identify a student, select a period, and select a reason for absence from a modifiable, drop-down pick list. The OSD Information Technology (IT) staff added an option for "a-ILI."

| School incentive
Each school receives $4000 per year to defray costs associated with IT support and effort by the attendance staff.

| Specimen archiving
Aliquots of all residual specimens are archived at À70 C at WSLH.

| Validation of influenza vaccination
We validate influenza immunization status for all ORCHARDS students using the Wisconsin Immunization Registry (WIR). 37   Annual levels of absenteeism were similar across all 5 years (Table 2).
We completed 1728 home visits for children with ARI. Children ranged in age from 4 to 18 years (mean AE std. dev. = 9.9 AE 2.5 years).

| DISCUSSION
In contrast to routine surveillance relying on MAI, ORCHARDS uses a community-based design. Most cases of influenza do not present for medical attendance 40 or result in hospitalization or death. 41,42 Influenza attack rates are much higher in school-aged children than for any other demographic group, 40 and influenza significantly contributes to school absenteeism. 43 This is reflected by the prominence of influenza

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1111/irv.12920.

DATA AVAILABILITY STATEMENT
The datasets generated and/or analyzed during the current study are not publicly available because the study is ongoing but may be available from the corresponding author upon reasonable request.