Legionellosis Surveillance System Evaluation - New Jersey, 2006-2013

BACKGROUND:  Cases of legionellosis have been on the rise all over the United States since its discovery in 1976.  In fact, the CDC estimates that the rate of reported legionellosis infections has increased 192% from 2000 to 2009 nationwide.  In New Jersey, rates of reported legionellosis have been regularly higher than national rates and have nearly doubled in the last 10 years as measured by the New Jersey Communicable Disease Reporting and Surveillance System (CDRSS).  Because of this continued increase, CDRSS was evaluated by the New Jersey Department of Health (NJDOH) to assess whether the system was fulfilling potential uses and to measure various attributes of surveillance systems.

METHODS:  The CDC guidelines for evaluating a public health surveillance system were used to summarize proficiencies and limitations in New Jersey’s surveillance and to make appropriate recommendations for improvements.  All attributes outlined in the guidance document were addressed, such as acceptability and stability; however, only three main attributes were analyzed for this surveillance system: data quality (as measured by percent missing of important variables), timeliness (as measured by median days between key status changes), and sensitivity (not quantified, but assessed by literature review).

RESULTS:  Data quality was found to be lacking, with four out of seven case-specific demographic variables missing significant amounts of data.  Timeliness was found to be acceptable, with a median of 10 days between case creation and case closeout, among other time intervals calculated.  In its ability to detect all real cases, CDRSS was believed to be lacking in sensitivity mostly due to factors outside surveillance, such as testing and clinical practices.  In its ability to detect outbreaks, CDRSS was believed to be sufficient in detecting outbreaks.  Since CDRSS is a well-established system for all reportable diseases, widely used, and secure, the flexibility, acceptability, simplicity, and stability were all known to be acceptable and were therefore not evaluated in depth.

CONCLUSIONS:  Overall, the system succeeded in its basic uses, such as capturing legionellosis disease trends and detecting outbreaks in a timely manner.  However, the system was found to have limitations in measuring risk factors and directing the most effective prevention methods within certain populations. Findings from the evaluation highlight what changes can be made to improve surveillance for cases and overall usefulness, which will be taken under consideration in the upcoming system redesign.