ELR Data Quality (Minnesota Department of Health)

Hi Sarah,

We are not working on ELR data quality for our project, but below are some resources/info I have previously come across that are hopefully helpful:

Best,
tiffany

Electronic_Laboratory_Data_Quality_and_the_Value_o.pdf 245 KB • Download

Thank you for these resources Tiffany, I'll look into them! 

Hi Sarah,
Sorry for the delayed response to your initial post! We at Maryland HIV Surveillance are somewhat working on ELR data quality for our project, by monitoring the format, volume, and distribution of laboratory results that are being reported to us and generating laboratory "report cards" to evaluate each reporting facility. However, we are also doing a lot of ELR data quality work outside of our AITT project, and we wanted to comment to Tiffany's post above that we are using the SaTScan software to detect drop-offs in our ELR, which has been interesting and useful for us to explore. For HIV in particular, we have used the SaTScan software to detect drop-offs in total ELR volume reported by facility, as well as monitoring the volume of specific HIV laboratory test types being reported monthly. We also have a thorough QA process and monthly log, as well as an ELR QA team that meets weekly to monitor reporting and data quality issues. We'd be happy to share more information with you directly offline, if you are interested!
Thanks,
Alyssa

Hi Alyssa, 

We would be interested in learning more about your approach using SaTScan and your overall ELR QA strategy as well. 

Thanks, 

Randi Rosack, TN Department of Health

I would like to here more about the approach using SaTScan as well.
Regards,
Ben Klekamp (VA)

I am interested, too. I have used SatScan in the past - for environmental health projects (to detect environmental contaminants across geographies). So, I'd like to see a different application and use of the software. Thanks!

Hi everyone, 
 
 Sorry for the delay in responding to your interest in the work the QA work that the Maryland HIV Surveillance team is doing. Regarding our general QA processes, Maryland has an Epidemiologist assigned to each lab/facility that sends reports to us. Each lab has a specific SAS program that was developed to put all records into a consistent format that can be uploaded to our surveillance system (eHARS). Each file is then reviewed by another epidemiologist before uploading. Information on each report, such as number of results in each report and received, QA and uploading dates, is added to a monthly log. This log allows us to keep track of what has been done, what needs to be done and keeps a record of any problems that need addressing. Our QA team meets every week to discuss any relevant topics or issues as they arise. All of our processes are described in a QA manual, which is updated regularly. We would be happy to share the QA manual offline if you were interested.

We have been working with SaTScan since New York City presented their work at CSTE last year. Tiffany included a link to that presentation above. Which was very helpful in establishing our usage of the software. Our utilization of SaTScan is essentially the same as NYC’s application. The integration of SaTScan and Tableau is to analyze the output of SaTScan easier. To do this we merged the input and output files before creating a dashboard that creates a timeline of ELR by facility and highlights the drop off period of ELR volume identified by SaTScan. This allows us to determine if follow up is necessary without investigating each facility by hand.

Thanks,
Veronica