Evaluation of the Los Angeles County Coccidioidomycosis Surveillance System from 1996 to 2013

BACKGROUND: Coccidioidomycosis (Valley Fever), endemic to southern California, is an infection resulting from the inhalation of airborne spores of Coccidioides immitis or Coccidioides posadasii.  Coccidioidomycosis has been a reportable disease in California since 1928, and is reportable within seven calendar days. We evaluated the Los Angeles County (LAC) Acute Communicable Disease Control Program (ACDC) coccidioidomycosis surveillance system.

METHODS: CDC’s Updated Guidelines for Evaluating Public Health Surveillance Systems were used to evaluate LAC’s coccidioidomycosis surveillance system from 1996 to 2013. The performance of the system as it relates to flexibility, stability, data quality, and timeliness was evaluated. We reviewed ability for the system to adapt to disease specific changes. LAC follows the current Council for State and Territorial Epidemiologists coccidioidomycosis case definition requiring clinical presentation and a single serology unlike the 1996 case definition. In 2005, LAC instituted an electronic laboratory reporting system (ELR) which automated lab-based data. Timeliness was evaluated using the difference between the case report date and the date of diagnosis.

RESULTS: Flexibility: The electronic form that captures attributes of the case definition can be changed and adopted with ease. One staff enters cases into the database. Stability: The results fields of many variables in the database vary in accuracy and completeness. Case confirmation relies on healthcare providers to send appropriate clinical documentation. After implementing ELR, cases significantly increased from 2004-2006. Without specific database variable definitions, subjectivity creates potential bias in the data entered or skipped. Timeliness: Among  6,984 reported cases including false and confirmed, the difference between the date reported and the diagnosis date ranged from one to 1,291 days (median= 20, missing=4,225). Date quality: Among 2,121 cases with hospitalization data, 36.6% of cases were hospitalized. Among 1,407 cases with fatality data, 5% of cases died.

CONCLUSIONS: The coccidioidomycosis case data form is editable and flexible. Due to changes in the surveillance protocol and data entry, the system has been inconsistent from 1981 to 2013. The database has become more complete showing greater stability with less opportunity to enter data in open-text format, data completeness via ELR, and consistency in personnel collecting and managing cases. Limited access to electronic medical records decreases timeliness of reporting. Timeliness may be reflective of lack of timely reporting by healthcare providers along with a need for clinical and laboratory confirmation and lack of public health resources. The fatality and hospitalization proportions may be an indicator of case underreporting and underdiagnosing.