How to solve healthcare’s identity challenge

Patient identity matching is one of healthcare’s most perennial challenges.

Accurate medical records are crucial for the continuity of care of patients. Many of us at Alcidion have worked in facilities that have medical records going back decades! The ability to access identified patient data is increasingly important as our understanding of diseases and genetics expands.

In the 90s, there was a growing movement to improve patient identification for better patient safety. The first Enterprise Master Patient Index (EMPI) was introduced with a vision to replace paper medical records and improve the accuracy of every patient’s identity and the availability of their information.

Today, the failure to identify patients and match them to intended treatment in care settings continues to lead to medication errors and wrong site procedures in healthcare systems across the world. In many cases, the challenges relating to patient identification have not changed.

For example, duplicate registrations, which occur when registration clerks re-register a patient who has already been provided with a unique identifier, continue to occur in healthcare facilities and are often impossible to remove from the health system. The causes of duplication are multiple and include inadequate searching; unconscious patients; and baby registration practice.

The technology exists to address this problem, so why does the healthcare sector still struggle with the accurate identification of patients? We believe there are three main reasons:

1. We ask humans to identify humans

Every human is prone to mistakes, particularly when it comes to spelling or pronouncing other people’s names. In healthcare, it is easy for verbal communication to be misinterpreted. For example, a patient says his name is “Lee” and the registration clerk records him as “Leigh”.

Technology can address misinterpretations in a couple of different ways, including Soundex searching, to search for alternative spellings of a name using the way the name is pronounced; dictionaries of common abbreviations such as Phil, Phillip and Philippa; and the searching of aliases, to address common scenarios such as maiden versus married names.

However, these methods still rely on the human to recognise the name in the list of returned records. Further, the list of records needs to be in an order that represents a likelihood of match so the desired record is more likely to be at the top of the list.

2. We assume patients are honest

In some cases, a patient might not be honest in their response to identification questions. For example, an Alzheimer’s patient brought to ED by first responder staff may not recollect their address. A patient in the middle of a mental health episode might cogently provide fake details. A patient could provide a relative’s details to avoid paying for care.

Registration staff in hospitals face these situations on a regular basis. There is no simple answer to these circumstances and the health systems must be designed to cater for these situations.

3. We use unintuitive systems

Health systems need to better control the registration process. Some systems allow a user to register a patient without searching the system or they return possible candidate results in a random order, making it more difficult to find the correct record.

Some systems also allow a user to search for a patient using one set of criteria and then register a new record with completely different values or allow an existing record to be edited with no limit to the fields changed. For example, a record based on a search for 85 year-old John Smith is registered as 19-year old Mary Jones.

Benefits of an EMPI solution

Modern EMPI solutions leverage sophisticated matching algorithms that address issues around misspellings, similar names or historic names. These solutions can provide probabilistic searching algorithms and result sets so that the most likely records are presented at the top of the result set.

By coordinating the demographic updates from all systems into a single record, patients only need to provide their change of address to the healthcare organisation once, not on every attendance. It is also easy to integrate EMPI features into third-party systems to streamline search and registration processes. In addition, EMPI solutions will provide a set of back end workflows to automate and guide the process required to identify and rectify duplicate registrations as they occur.

Approaches for an EMPI solution

When implementing an EMPI solution, healthcare organisation must first define the role of an EMPI in their business process and system architecture. Some government health departments leverage EMPI to provide backend linkages of records from participating systems, which enables them to relate activity across multiple organisations, without changing how patient identification is managed.

Other health departments have implemented their EMPI solutions in a passive mode. They are called “passive”, as while the EMPI receives feeds from all systems and updates its “Single Best Record” view of the patient, the EMPI does not send those updates out to participating systems. This approach is favoured when looking at population behaviours and patterns.

The alternate approach is the use of a single patient identifier across a whole jurisdiction. For example, the New Zealand National Health Identifier (NZNHI), the Australian Individual Health Identifier (IHI) and United Kingdom’s NHS Number. These identifiers link the patient across the enterprise with a common identifier, streamlining communication and patient information sharing across providers.

It is also important that electronic health records have unified identification for health professionals and care givers. The ability to unequivocally identify those providers involved in providing care and support services to patients is both a safety and a medico-legal requirement. However, most organisations maintain separate provider directories in each of their systems. This can lead to administration overhead when updates need to be made across multiple systems.

One health jurisdiction in Australia has gone as far as implementing a searchable services directory, where the relationships between services, health professionals and locations are made available to authorised users. This foundation capability streamlines the referral process from primary care to specialist services, directly contributing to improved and timely patient care.

At Alcidion, we work with multiple healthcare organisations and jurisdictions to improve and establish streamlined identification processes. Some of this work has been progressed with our partner, NextGate Solutions, who provide the world-leading EMPI in MatchMetrix.

Paul Marriott is a Product Manager at Alcidion