Patient identification errors happen more often than they should. Clinicians may confuse two patients for one another or pull the wrong chart for a patient whose name sounds similar to another. These mistakes can be costly and deadly when they result in scenarios like giving a patient the wrong kind or amount of medication or pulling an advance directive for a patient from the wrong health record.
Though most healthcare workers doubt they would ever misidentify a patient, no one is immune from these errors, including physicians, nurses, lab technicians and pharmacists. Up to 12 percent of hospital medical records are duplicates, and up to 96,000 medical records in the average electronic medical record (EMR) system refer to a patient with another existing medical record, according to biometric identification provider RightPatient. In southern California alone, regional medical provider Kaiser Permanente has more than 10,000 records of people named Maria Gonzales.
Inaccuracies tied to an incorrect patient identity occurs in up to 14 percent of medical records, reports the Healthcare Information and Management Systems Society (HIMSS). Though many facilities rely on bar-code scanners to confirm the identity of patients and EHR systems that alert clinicians to duplicate names, none of these are foolproof solutions.
How Interoperability Factors In
Patient identification errors were listed among the top 10 patient safety concerns by ECRI Institute, a nonprofit patient safety advocate group. A recent report by ECRI, which examined more than 7,600 cases of patient errors recorded between January 2013–June 2015 at more than 181 facilities, found that idiosyncrasies of EMRs contributed to the problem. EMRs may not recognize minor variations in name spellings, which can lead to duplication of patient files and blending of data for two individuals. Similar birthdates and zip codes for individuals may also trip up the system. While some systems have criteria that matches patients to past records, these auto merges can cause even more problems if the matching criteria or algorithms aren’t specific enough—or the applications or staff managing the master patient index (MPI) are not closely monitoring and managing these merges.
Data sharing among disparate health IT systems also causes confusion, according to the report. A patient may have multiple records across several different healthcare systems; yet not all of these records include the same identifying information or can transfer that data accurately to another system.
Human error is also to blame. Doctors, nurses, technicians and registrars often have different standards for collecting information—they may or may not ask a patient how to spell his or her name or street address, for example, or ask for other identifying info such as a middle initial, birthdate or phone number. Nearly 13 percent of identification errors highlighted in the ECRI report happened during patient registration, and 22 percent occurred during procedures and tests. At times, patients’ wristbands were missing, unreadable or not even checked at all, the report noted.
One key to reducing patient identification errors is implementing patient matching algorithms and technology. Another is developing an enterprise master patient index (EMPI), an interoperable engine that matches and aggregates data identifying patients across information systems.
A universal patient ID, much like an automobile’s vehicle identification number (VIN), would also help providers locate and access the correct patient record, asserts healthcare writer Dan Munro. But more immediately, ECRI encourages providers adopt a standard means of patient identification, such as photographs with patient files. Clinicians and hospital executives should also make avoiding identification errors a bigger priority when training staff, the report asserts.
Explore more about how to overcome challenges to interoperability of patient data in our white paper.