HLTH 4200 Week 2 Surveillance Note Example

School Absence Records as a Signal for Influenza-Like Illness in the Brookfield Unified School District: How the Data Are Collected, What They Can Signal and Who They Never Count

Student Name

Walden University

HLTH 4200: Principles of Epidemiology

Instructor Name

Month Day, Year

School Absence Records as a Signal for Influenza-Like Illness in the Brookfield Unified School District: How the Data Are Collected, What They Can Signal and Who They Never Count

Purpose

This note describes one data source the Brookfield County Health Department, a composite local agency, is considering for influenza surveillance: daily absence records from the twelve buildings of the Brookfield Unified School District, also a composite. It describes how the records are collected, what an illness-related absence can signal about influenza-like illness in the community and which people and illnesses the method never counts. Public health surveillance means gathering and examining health data continuously and on a fixed plan so that the results can guide action (Centers for Disease Control and Prevention, 2012), and any data source should be described in those terms before it is used.

How the Records Are Collected

Each of the district's twelve buildings, eight elementary schools, three middle schools and one high school, enrolls between 350 and 1,600 students, for about 9,200 students in total. Attendance is taken by the homeroom or first-period teacher by 9:15 each morning in the district's student information system. When a parent calls the attendance office, the clerk records a reason code: illness, medical appointment, family or religious reason, or other. Students absent without a call are recorded as unexcused until a reason arrives; many reasons are never recorded.

For surveillance, the district would send the health department, each school day by noon, a file with the count of students enrolled and the count absent for each reason code in each building. No names or student identifiers would be included.

Table 1

Data Elements in the Daily File

The health department would calculate the percentage absent for illness in each building and across the district each day.

What an Illness Absence Can Signal

Influenza spreads readily among school-age children, and children often become ill early in a community's influenza season. A rise in illness absences across several buildings at once could therefore signal that influenza-like illness is increasing in the community, possibly before rises appear in emergency department visits or laboratory reports. The signal is most useful when it is compared with the same buildings' usual levels for the same time of year, since absence rates vary with weather, holidays and testing schedules.

Evidence on how well this works is mixed. An evaluation in New York City examined three years of daily school attendance data and found moderate increases in absenteeism among children during peak influenza season, suggesting some value for detecting large citywide epidemics (Besculides et al., 2005). The same evaluation found school-level data too noisy to detect localized outbreaks reliably, and the city decided not to adopt prospective monitoring of total absenteeism. That evaluation used total absence, not absence coded for illness; this note proposes using the illness code, which may be more specific but depends on parents calling and clerks coding consistently.

What the Method Never Counts

A surveillance data source is defined as much by what it misses as by what it captures. Several groups and illnesses fall outside this one entirely.

Sick children who attend school. A child with influenza who comes to school, perhaps because a parent cannot stay home, is counted as present. In years when many families cannot afford time off work, illness absences may undercount illness most in the neighborhoods with the most need.

Sick adults. Teachers, staff and parents do not appear in student absence data. If an influenza strain affects older adults more than children, the signal may be weak or late.

Children not enrolled in the district. Children in private schools, homeschooled children and children under school age are outside the data.

Illness on days without school. Weekends, holidays and summer produce no data. A rise in illness that begins over a holiday break would not appear until school resumes, and the first day back may show an artificial spike as illnesses that began during the break are recorded together.

Absences with no reason recorded. Unexcused absences with no call may include many illness absences. If the share of absences without a reason changes over time, the illness count changes for reasons unrelated to illness.

Illness absences coded as something else. A child kept home with a fever but recorded as a family reason is missed. Coding practices may differ between buildings and clerks.

Routine absences coded as illness. Some parents report illness for absences with other reasons, which adds noise unrelated to infection.

Judging the Data Source

Guidelines for evaluating surveillance systems direct attention to attributes such as simplicity, timeliness, data quality, representativeness and sensitivity (German et al., 2001). Against these, the absence data are simple and timely: they already exist and could arrive by noon each school day. Their weaknesses are representativeness, since they cover only enrolled children on school days, and data quality, since the illness code depends on parent calls and clerk coding. Their sensitivity for detecting a citywide or countywide rise may be moderate; for detecting an outbreak in a single building, it may be low, as the New York evaluation found.

How a Signal Would Be Judged

Before the first season, the department should decide what counts as a signal. One approach is to compare each day's district-wide illness absence percentage with the average for the same calendar week in the two previous school years, flagging days that exceed that average by a set margin for two consecutive school days. A building-level flag would require a larger margin, because a single classroom's illness can move a small building's percentage sharply. Writing these rules down in advance prevents the temptation to call any rise a signal after the fact.

Recommendation

The health department should use illness absences as one supplementary signal, not a stand-alone alarm. It should establish a baseline from at least two prior school years, track the share of absences with no reason code alongside the illness code, and compare rises against emergency department and laboratory data before acting. After one season, it should evaluate whether the signal added anything those sources did not already show.

References

Besculides, M., Heffernan, R., Mostashari, F., & Weiss, D. (2005). Evaluation of school absenteeism data for early outbreak detection, New York City. BMC Public Health, 5, Article 105. https://doi.org/10.1186/1471-2458-5-105

Centers for Disease Control and Prevention. (2012). Principles of epidemiology in public health practice: An introduction to applied epidemiology and biostatistics (3rd ed.). U.S. Department of Health and Human Services. https://archive.cdc.gov/www_cdc_gov/csels/dsepd/ss1978/index.html

German, R. R., Lee, L. M., Horan, J. M., Milstein, R. L., Pertowski, C. A., & Waller, M. N. (2001). Updated guidelines for evaluating public health surveillance systems: Recommendations from the Guidelines Working Group. MMWR Recommendations and Reports, 50(RR-13), 1-35. https://www.cdc.gov/mmwr/preview/mmwrhtml/rr5013a1.htm