Although duplicate medical record numbers (MRNs) certainly are problematic, registering a patient on another person’s MRN is dangerous.

“This is particularly true with the implementation of EMRs as this can lead to medical decisions that are based on a different patient’s history,” says Shannon Haager, director of patient access services at The Ohio State University Wexner Medical Center in Columbus.

For this reason, Haager says, “We are much more uncompromising about the selection of a wrong MRN than with the creation of a duplicate.”

If, after using several search techniques, registrars aren’t absolutely positive their patient matches an existing MRN, they create a new record.

“The most common reason for this is two people with the same name and other data points are not validated,” Haager says.

A registrar assumes he or she has the correct “Tabitha Jorgenson” because only one comes up in the system, for instance.

“But it’s a brand new patient, raising doubt,” Haager notes.

Researchers found that 98.81% of individuals were uniquely identified using first name, last name, and date of birth while examining a database of more than 85 million individuals.1 However unlikely, it’s still possible to misidentify patients, even with those identifiers.

“We have several sets — at least five that I know of — where two people have the same name and date of birth,” Haager says.

Haager finds that talking with staff about what happens when the wrong MRN is used — “specifically, how dangerous it can be — is the most effective tool for reducing misspellings.”

Protocols to stop duplicate medical records were put into place recently at the centralized scheduling department at The Ohio State University Wexner Medical Center.

“We work directly with patients to verify multiple pieces of information,” explains patient access director Brooke Bellamy. This confirms whether it’s an existing patient in the system, or if a new account needs to be created.

“We have also standardized our registration process,” Bellamy adds. “We are confirming and collecting accurate data for patients for every appointment we schedule.”

Patient access quickly and easily identify any records that could match the patient, so a duplicate record isn’t created. The current processes still rely on the information given by patients, though.

“This allows for errors to be made,” Bellamy acknowledges.

Patients don’t always feel comfortable giving or confirming their Social Security numbers, one of the few pieces of information that does not change, over the phone. Patient names, aliases, addresses, and phone numbers often change.

“It leaves room for errors resulting in duplicate records,” Bellamy notes.


  1. Zech J, Husk G, Moore T, Shapiro JS. Measuring the degree of unmatched patient records in a health information exchange using exact matching. Appl Clin Inform 2016;7:330-340.


  • Shannon Haager, Director, Patient Access Services, The Ohio State University Wexner Medical Center, Columbus. Phone: (614) 293-7670. Email:

Study IDs Root Causes of Duplicate Records

Incorrect middle names, Social Security numbers, misspellings, and first, middle, and last names in the wrong order were the most common causes of duplicate records, according to a study on a database of 398,939 patient records with confirmed duplicates.1

“The findings weren’t a surprise to me as we had studied this back in 2007 and 2008,” says Beth Haenke Just, MBA, the study’s lead author, noting the only improvement was the accurate capture of the patient’s date of birth. “I believe this is due to the large emphasis over recent years on positive patient identification.”

Just suggests that patient access departments incorporate these findings into their training programs. “If patient access staff understood more clearly the top data capture challenges and their impact on duplicate record creation, their data capture quality would increase.”

Just views capturing the patient’s legal name accurately as a top challenge. “Nearly 50% of all duplicate records have a discrepancy in the last or first names, some with both,” she says. Nicknames and misspellings are common, as are last name changes for women and children due to marriage or divorce. Just recommends these practices:

  • Query the patient or parent about prior names used;
  • Obtain a legal document from the patient, such as a driver’s license, and use that for entering the patient’s name;
  • Ask the patient “Have you ever been here, or anywhere in the health system, before?” If the patient says “yes,” but the registrar does not find the patient when searching the database, ask, “Have you ever been here under another name?”
  • Ask for the patient’s full middle name. “This ensures there are enough data points captured on each record to uniquely identify the patient,” Just says.


  1. Just BH, Marc D, Munns M, et al. Why patient matching is a challenge: Research on master patient index (MPI) data discrepancies in key identifying fields. Perspect Health Inf Manag 2016 April 1.