Communicable disease surveillance is essential for protecting public health (1,2). The Utah Department of Health (UDOH), Bureau of Epidemiology (BOE), like many other state health departments, relies heavily upon disease reports from local health departments (LHDs), laboratories and clinical providers to rapidly identify and respond to disease trends and outbreaks so that additional illness might be prevented (3,4). The Institute of Medicine recently cited a “glaring need” to bolster disease reporting to prepare for the threat of bioterrorism and to appropriately deal with emerging pathogens (5). Previous evaluations of notifiable disease surveillance in various states have found that illnesses are not only substantially under-reported, but that reports are often submitted late (6,7).

In an effort to determine the timeliness of disease reporting and the level of compliance to reporting standards for those diseases, the BOE, Surveillance Program examined the median reporting times for ten reportable communicable diseases in Utah for the 1995-1999 period.

Methods

Utah Communicable Disease Registry: The Communicable Disease Rule (Utah Rule R386-702) mandates that clinicians and laboratories report confirmed or suspect cases of reportable diseases to either the BOE or the LHDs. Electronic reporting was instituted in 1998; therefore, data from the 1995-1997 period were selected to compare to the initial period of electronic reporting.

Data sources: Data was used from the National Electronic Telecommunication System for Surveillance (NETSS), developed with EpiInfo 6 software by the Centers for Disease Control and Prevention. The NETSS contains data fields such as “event date” (earliest known date of disease onset) and “report date” (date in which an event was reported to either the BOE or the LHDs). These two data fields were used in the analysis.

Selection of diseases: Ten reportable diseases that must be reported to either the BOE or the LHDs within seven calendar days from the time of diagnosis were selected. The diseases were selected because they had the greatest frequency in the state and represented 30 or more cases per year, allowing for greater statistical power. The selected diseases were campylobacteriosis, E. coli O157:H7 infections, giardiasis, hepatitis A, hepatitis B, influenza, bacterial meningitis, pertussis, shigellosis, and salmonellosis.

Statistical Analysis: Annual median reporting times for these diseases were calculated for years 1995 to 1999 using EpiInfo 6 software. Data was available for a statistical analysis from years 1990 to 1999. However, the 1990-1994 period was not chosen for the analysis because those data were collected using different reporting requirements and data collection techniques. The median values of the 1995-1999 period were compared to establish the appropriate measure for timeliness of reporting. The 25 and 75 percentiles were also calculated to establish whether or not changes in reporting times were statistically different from year to year. Additionally, a linear regression was applied to each disease data set to determine the relationship between reporting times each year in the 1995-1999 time period.

The data was positively skewed due to outliers in excess of hundreds of days for several cases. Outliers, which affect the mean and not the median, rendered the median values a better representation of the results in this study. Therefore, since data for this analysis was not normal, a non-parametric evaluation was conducted for this study.

A linear regression test applied to each disease data set indicated that the reporting times for each of the ten diseases decreased and, therefore, improved during the 1995-1999 period. However, none of the improvements were statistically significant when differences in variability were compared for each year included in the study. The results of the study are presented in Figure 1.

The range of median reporting times in 1995 was as low as nine days for influenza and as high as 30.5 days for pertussis infections. In 1999, that range narrowed from a median of eight days to report influenza infections, to 24.5 days to report pertussis infections (table 1). (One of the criteria for a report of pertussis to meet the CDC case definition is a history of cough lasting at least 14 days. The cough onset date is the date entered in the NETSS as the event date for confirmed pertussis cases. The apparent delay in reporting pertussis cases may be in part due to the unique criteria of defining pertussis.) Though this range narrowed during the study period, all of the minimum median values were still greater than seven days, the time before which all of these diseases should be reported to either the BOE or LHDs. While improvements have been made regarding promptness of reporting, these range values are representative that less than half of the cases corresponding to the ten diseases used in the analysis are reported within the required time period.

Timely reporting of communicable diseases renders the BOE able to promptly perform case investigations and contact tracking and, therefore, aids in preventing additional illness.

The analysis showed an observable decrease in median reporting times for each of the ten diseases analyzed. This analysis presented an interesting dichotomy between actual reporting times and the required reporting times. The assessment illustrated a decrease in median reporting times for all of the diseases analyzed. However, all median reporting times were still more than seven days, the time frame in which the Communicable Disease Rule mandates that these 10 diseases should be reported to the BOE or LHDs. Therefore, while substantial improvements have been made regarding promptness of reporting, less than half of the cases analyzed in this study are actually reported within the required time period.

Improvements in reporting times from 1995 to 1999 are possibly due to several factors such as more efficient surveillance methods and electronic reporting. Electronic reporting between LHDs, selected laboratories, and the BOE was instituted in 1998. A recently published study stated that electronic reporting systems tend to capture more reports and improves timeliness of reporting by nearly four days (8). By 2000, all of the twelve LHDs in Utah were reporting electronically. Future evaluations of reporting times will hopefully result in a further reduction of mean and median reporting times.

This assessment examined the median reporting times for cases of campylobacteriosis, E. coli O157:H7 infections, giardiasis, hepatitis A infections, hepatitis B infections, influenza, bacterial meningitis, pertussis, salmonellosis, and shigellosis during the period 1995-1999. This analysis demonstrated an observable decrease in reporting times for each of the diseases. Additionally, although the median reporting times decreased for the ten diseases, they were still all longer than required by the Utah Communicable Disease Rule. All suspect and confirmed cases of the 10 diseases analyzed should be reported within seven days to effectively prevent further illness.

Additional assessments should be performed to determine mean and median reporting times at the level of each LHD to assess how they are affecting statewide data. There may be several factors affecting the timeliness of disease reporting such as electronic reporting, understaffed LHDs, apathy, lack of time in clinical providers’ offices, and as mentioned earlier, the criteria for defining pertussis. Further study should focus on reasons for both prompt and late reporting times in the state in order to implement proper interventions to improve the timeliness of disease reporting.

1. Chorba, T. L., Berkelman, R. L., Safford, S. K., Gibbs, N. P., Hull, H. F. (1990) Mandatory reporting of infectious diseases by clinicians. MMWR Morb Mortal Wkly Rep. 39(RR-9):1-17.

8. Effler, P., Ching-Lee, M., Bogard, A., Ieong, M., Nekomoto, T., Jernigan, D. (1999) Statewide system of electronic notifiable disease reporting from clinical laboratories, comparing automated reporting with conventional methods. AMA. 282: 1845-1850.

Return to Top