Changes in Data Completeness with Electronic Laboratory Reporting (ELR) Implementation in Detroit City, Michigan

Background:
Michigan, including Detroit City, has a significant percentage of unknown chlamydia and gonorrhea cases reported with unknown race, zip code, age, address, and sex.

Objective:
To assess changes in completeness of several variables following the implementation of ELR and plan for improved data quality transmission at a major medical center in Detroit City.

Method:
Gonorrhea and chlamydia cases reported between January 1, 2005 and June 30, 2007 with a residence in Detroit City were exported from the system. The percent of missing or unknown data was examined by quarter for race, age, zip code, sex, and street address. Significance was assessed using Cochran-Armitage Trend Test.

Result:
For chlamydia cases, there was a statistically significant increase in completeness for age, zip code, and address. There was no significant change for race and a significant increase in missing sex data. For gonorrhea cases, there was a statistically significant increase in completeness for race, age, zip code, and address. There was no significant change in sex completeness.

Conclusion:
ELR has significantly improved completeness levels for many variables, except for sex. This is presumably because in the past sex of the case could be imputed from a paper report by looking at name or site of specimen.

Implications:
Leveraging ELR data from major STD reporters to include required demographic fields could have a significant impact on overall completeness levels and thus enable better epidemiological characterization of the at-risk population.