Our Approach
EMPI Consulting
MPI Cleanups
 Data Analysis
Software Solutions
Patient Identity Consulting
HIPAA Privacy and Security
Data Integration Services
Cerner Millennium Services
Meet Our Consultants
Resource Center
|
 |
 |
Duplicate Medical Records Analysis
Most Health Information Management systems use some form of duplicate detection algorithms to help identify duplicate medical records within a database. These embedded algorithms typically use some form of Deterministic, Probabilistic, and Rules Based logic to identify duplicate records. These algorithms tend to identify duplicate medical records that match "exactly". They miss a large percentage of duplicates where the data is just off slightly. They also have a very high False Positive rate. While Just Associates will perform a cleanup using any duplicate reports our clients prefer, we found that a data analysis conducted with the more sophisticated matching algorithms provided by our partner, Netrics, yields a more accurate analysis. As a result, our efforts to identify and resolve duplicate mpi records are more effective.
Just Associates utilizes error-tolerant record matching technology to detect duplicate patient records within a single, and across multiple, MPI / EMPI databases. Just Associates has performed scores of MPI data analyses and has examined over a hundred million patient records, accurately identifying duplicate medical records and shared MRNs despite working with inaccurate, incomplete, inconsistent, and unreliable patient demographic information. Our services range from delivering simple analysis reports to providing complete, turnkey, on-site reconciliations performed by experienced HIM professionals.
Intelligent Search
Just Associates' duplicate record search uses Netrics' proprietary "Bipartite Graph Matching" (BGM) algorithms to compare database records on a sliding scale. Netrics mathematically models the human notion of similarity to accurately measure how much a record has in common with each record in the EHR or MPI database. This flexible approach enables us to easily overcome impure data, data entry errors, mis-fielded data, and common variations in patient demographic information.
Duplicate Record Matching
Duplicate record matching technology builds on BGM by adding a Machine Learning component which learns by example and develops an understanding of what qualifies as a "match" in the context of your data and your needs. The program deals automatically with irregularities like missing data and blank fields that stop other matching approaches cold. Netrics' Machine Learning solution outperforms rules-based and probabilistic matching approaches, and, unlike them, does not require expert tuning or recalibration any time your database changes significantly.
Overlaps / Overlays
In addition to performing an analysis for duplicate records, your database can also be analyzed for Overlaps and/or Overlays. An Overlap occurs when a patient has more than one medical record number assigned across more than one database. An Overlay occurs when one patient record is overwritten with data from another patient's record.
Reports
After the data is analyzed, the results are compiled into various reports. Potential duplicate pairs are weighted and grouped for more efficient verification. Summary Statistical Reports and graphs display the distribution of potential duplicates across variables. Data Integrity Reports summarize the quality of the data within the database by showing the number of blank, invalid or default values contained in key demographic fields.
Our Experience
Our staff has experience using all of the duplicate detecting methodologies available today. In our experience, a data analysis performed using the advanced algorithms provided by our partner, Netrics, produces a better result. More true duplicate records are identified and at the same time the false positive rates are exceptionally low. A better data analysis leads to a more complete cleanup.
Click here to visit Netrics |
|
