HCI 5073 Module 1 Surveillance System Assessment Example

Heat Illness Within a Day: Describing and Assessing a County's Syndromic Surveillance of Emergency Visits With the CDC Surveillance Evaluation Attributes

Student Name

American College of Education

HCI5073: Public Health Informatics

Module 1 Assignment

Instructor Name

October 5, 2026

Purpose of the System

Picture a composite Midwestern county health department that needs to know, during a heat wave, whether people are becoming ill from the heat, where and how many, fast enough to change its response the same day. Death certificates and hospital discharge records arrive weeks or months later. Syndromic surveillance, which uses data from emergency department visits collected in near real time, offers a faster signal. Nationally, the National Syndromic Surveillance Program reports that most U.S. emergency departments now send data to the program, often within 24 hours of a patient's visit (Centers for Disease Control and Prevention [CDC], 2026).

This paper describes how the county uses syndromic data to monitor heat-related illness and assesses the system against the attributes recommended in CDC's guidelines for evaluating public health surveillance systems. The goal is to understand what the system does well and where it needs improvement before the next heat season.

How the Data Flow

The data begin at triage. When a patient arrives at one of the county's four emergency departments, the triage nurse types the reason for the visit in the patient's words, such as dizzy after working outside, and later clinicians add diagnosis codes. The hospitals' electronic health records automatically send a message for each visit, containing age group, sex, residential ZIP code, chief complaint, triage notes and diagnosis codes, to the state health department's syndromic system, which in turn feeds the national platform. The state applies a standard query that flags visits as heat-related if the chief complaint contains terms such as heat, sun stroke or overheated, or if a heat-related diagnosis code is present. County epidemiologists see the results on a dashboard, updated daily, with counts and rates by day, age group and ZIP code.

Understanding the flow matters because each step shapes what the data can show. A triage nurse who writes weakness rather than heat will cause a heat illness to be missed, and a diagnosis code added days later will change the count retrospectively.

The Evaluation Framework

German et al. (2001), writing for CDC's guidelines working group, recommended that a surveillance evaluation describe the system's purpose, operation and resources and then assess its usefulness and nine attributes. Some concern how the system operates (simplicity, flexibility, stability and acceptability), some concern how well it finds cases (sensitivity and predictive value positive), and the rest concern its data (quality, representativeness and timeliness). The attributes often trade off against one another, so the aim is not to maximize all of them but to judge whether the balance fits the system's purpose. For heat surveillance, timeliness and sensitivity during heat waves matter most, while perfect accuracy for each individual visit matters less.

Assessment Against the Attributes

Timeliness is the system's greatest strength: most visits appear on the dashboard within a day, which allowed the department last summer to see a rise in heat-related visits among adults over 65 in two ZIP codes on the second day of a heat wave and to redirect outreach. Simplicity and stability are good, since the system runs automatically on existing records with little staff effort and experienced no outages during last summer's warnings. Flexibility is also good; the same data stream supports surveillance of overdoses, influenza-like illness and injuries.

Sensitivity and predictive value positive are weaker. A manual review of 300 visits during a heat wave, comparing the query's results with clinicians' full notes, found that the query identified about 68% of visits that clinicians attributed to heat, missing many older patients whose complaints were recorded as weakness, falls or confusion. Of visits the query flagged, about 85% were truly heat-related. Data quality varies by hospital; one emergency department sends triage notes but not diagnosis codes for several days. Representativeness is limited because the system sees only people who reach an emergency department, not those who die at home or are treated elsewhere, and older adults living alone are likely underrepresented among visits relative to their risk. The system is fast and reliable, but it is fastest at counting the heat illnesses that are easiest to recognize.

Usefulness

The guidelines ask whether a system actually contributes to action. The county's heat surveillance did: it triggered targeted outreach, informed daily situation reports to the emergency management office and provided the counts used in the department's after-action review. It is less useful for setting warning thresholds. Research using hospital data across 1,617 U.S. counties found that heat-attributable hospitalizations begin at moderately hot heat index values, in some regions below the alert criteria the National Weather Service used during the study period, and that locally specific evidence can improve heat warning systems (Vaidyanathan et al., 2019). The county's syndromic data could support such local analysis, but it would need several years of consistent data and a more sensitive case definition.

Acceptability and the People Who Run It

Acceptability, in the guidelines' sense, is the willingness of people and organizations to take part in the system, and it deserves a separate look because the system depends on staff who are not public health employees. Hospital information technology teams maintain the feeds, and they respond to data problems only when the county or state contacts them. Interviews with the four hospitals' interface staff suggested that participation is secure because it satisfies federal requirements for exchanging data with public health, but interest in improving data quality is uneven, and fixes compete with the hospitals' own priorities. Triage nurses, whose free-text notes drive the query, are largely unaware that their wording is used for surveillance. At the county level, the epidemiologist who monitors the dashboard values the system but reports that summer workloads leave little time to investigate unexpected spikes, and emergency management staff said the daily report is most useful when it arrives before their morning briefing. These findings matter for any improvement. A revised case definition or new data field will succeed only if the hospitals are willing to implement it and the people recording the data understand why it matters, so the system's weakest link may be engagement rather than technology.

Resources and Governance

The system's costs are mostly borne upstream: hospitals pay for their electronic health record interfaces, and the state maintains the syndromic platform with federal support. At the county level, the system uses about a quarter of one epidemiologist's time during the summer to monitor the dashboard, run queries and write daily reports, plus occasional analyst time for validation reviews. Governance rests on a data use agreement between the state and county that defines which fields the county can see, prohibits attempts to identify individual patients and requires aggregation of small counts in public reports. Those rules protect privacy but also limit analysis: the county cannot follow a patient from an emergency visit to a hospital admission, and it cannot see visits by county residents to hospitals in neighboring states. Any improvement plan must work within, or seek to amend, that agreement.

Conclusions and Next Steps

The county's syndromic surveillance of heat illness is timely, simple, stable and useful for real-time response, which fits its primary purpose. Its main weaknesses are moderate sensitivity for heat illness in older adults, uneven data quality across hospitals and limited representativeness of the people most at risk. The next module will analyze data quality and interoperability more closely, and later modules will consider how a revised case definition, better hospital reporting and linkage with other data sources could address those weaknesses.

References

Centers for Disease Control and Prevention. (2026, September 1). About NSSP. https://www.cdc.gov/nssp/php/about/index.html

German, R. R., Lee, L. M., Horan, J. M., Milstein, R. L., Pertowski, C. A., Waller, M. N., & Guidelines Working Group Centers for Disease Control and Prevention. (2001). Updated guidelines for evaluating public health surveillance systems: Recommendations from the Guidelines Working Group. MMWR Recommendations and Reports, 50(RR-13), 1-35.

Vaidyanathan, A., Saha, S., Vicedo-Cabrera, A. M., Gasparrini, A., Abdurehman, N., Jordan, R., Hawkins, M., Hess, J., & Elixhauser, A. (2019). Assessment of extreme heat and hospitalizations to inform early warning systems. Proceedings of the National Academy of Sciences, 116(12), 5420-5427. https://doi.org/10.1073/pnas.1806393116