By Megan Pruente, MPH, RHIA, Director, Professional Services, Harris Data Integrity Solutions
Twitter: @PatientMatching
While the current state of patient matching is best described as tumultuous, progress toward a solution is being made thanks to efforts of stakeholders from across the spectrum. In the past year alone, we saw Section 510 temporarily removed from the U.S. House and Senate Labor, Health and Human Services, Education, and Related Agencies (Labor-HHS) appropriations bills – the first time the Senate had done so in 20 years – and Office for the National Coordinator (ONC) issued its Project US@ Technical Specification for patient addresses and the ONC-AHIMA Companion Guide.
Patient ID Now also issued its Framework for a National Strategy on Patient Identity: A Proposed Blueprint to Improve Patient Identification and Matching to guide creation of a national strategy for patient identification to ensure accurate patient matching and protect patient safety. In it, the coalition noted that reliable patient matching is vital to patient care and essential to the advancement of interoperability and to support the access, exchange, and use of electronic health information. Furthermore, per the framework, “…inaccurate, incomplete, or inconsistently formatted demographic information in patients’ records have been shown to cause major adverse patient outcomes, including death.”
Exacerbating the patient matching problem is the rapid growth of the volume of patient data being aggregated and shared by healthcare organizations and patients themselves. In the absence of standards and an improved patient matching strategy, the impacts of healthcare’s patient matching problem will only worsen, further endangering lives and impacting the financial stability of healthcare organizations nationwide.
Broad Impacts
The financial impacts of patient misidentification are significant, costing the average healthcare facility $17.4 million per year in denied claims and lost revenue and the healthcare system as a whole more than $6 billion annually. Black Book Research also found that the expense of repeated medical care due to patient misidentification costs an average of $1,950 per inpatient stay and more than $800 per emergency department visit.
According to recent survey from HIMSS and Patient ID Now, healthcare organizations spend an average of 109.6 hours per week resolving patient identity issues. Over half spend 21-80 hours per week and have an average of 10 full time employees dedicated to patient identity resolution – despite nearly all participating organizations reporting that they had a unique patient identifier in place. More than one-third of responding organizations reported spending more than $1 million annually on identification resolution while just 18% said they spend less than $250,000 a year.
The impact of poor patient matching isn’t just financial. It is also a significant contributor to an average duplicate record rate that runs as high as 18% in the typical facility. Further, as many as 20% of all records are incomplete, forcing clinicians to provide care based on a potentially incomplete picture of their patients.
As a result, according to Patient ID Now’s framework: “Medications are prescribed for patients lacking a complete medical history in their record; allergies are missed, diagnoses are lost or delayed, and duplicative tests are ordered. The problem of patient misidentification is so dire that one of the nation’s leading patient safety organizations, the ECRI Institute, named patient misidentification among the top ten threats to public safety.”
The majority (70%) of respondents to the HIMSS/Patient ID Now survey concurred that troubles managing patient identities results in duplicative or unnecessary testing or services. Further, 67% of respondents said a lack of clear patient identities put their organizations at a higher risk for fraud, and 71% said it created identity verification and eligibility issues that made member enrollment and patient admission unnecessarily difficult. Finally, 78% percent indicated that inconsistent identity data complicates workflows, such as patient matching and eligibility.
Patient matching issues also impact response to public health emergencies, as was seen throughout the COVID-19 pandemic. Missing data, including an estimated 40% of demographic data missing from commercial laboratory test feeds for COVID-19, hindered contact tracing, vaccination, and public health reporting efforts. Patient ID Now also noted reports of vaccination registrations causing thousands of duplicate records within a single system, which cost some hospitals and health systems at least $12,000 per day to correct. There were also reports of vaccination sites being denied additional vaccines because inaccurate patient record systems showed patients as not having received previously administered vaccinations.
Industry trends are also to blame for patient matching issues. Seventy-seven percent of respondents to the HIMSS/Patient ID Now survey reported that electronic health record (EHR) migrations or facility acquisitions contributed to patient identity or duplication issues, and 71% indicated that portals allowing patients to self-schedule and/or register contributed to an increase in both identity issues and duplicate records.
Solving the Patient Matching Problem
Eliminating the patient matching problem will take a muti-stakeholder approach, starting with finally removing the outdated Section 510 from the Labor-HHS appropriations budget to allow HHS to explore the feasibility of adopting a unique patient identifier. But while this would set the stage for adoption of a national strategy for patient matching, it is only the beginning. What is also needed is a collaborative effort between the private and public health sectors, as Patient ID Now did when developing its framework, which serves as an excellent blueprint for improving identification and matching.
To eliminate matching errors, the framework sets forth recommendations for a national strategy that:
- Minimizes errors by improving matching rates across multiple scenarios, including addressing duplicates, overlays, and overlaps.
- Provides guidance on the matching and identity resolution process.
- Provides guidance, benchmarks, and standards for calculating error rates across health IT systems and organizations.
- Identifies performance measures, for example minimum acceptable levels of accuracy.
- Leverages work already underway at the federal level, including aligning with guidelines provided by the National Institute of Standards and Technology.
- Develops, disseminates, and conducts patient identification and matching training and encourages testing, evaluation, and optimization as appropriate.
The framework also strongly advocates for a standards-based approach that aligns with ongoing interoperability efforts. To that end, Patient ID Now recommends a national strategy that:
- Defines the minimum standardized data set needed for patient identification and matching, including adoption of a common set of specific demographic fields or data elements and a common set of standards for such data elements.
- Facilitates ongoing collaboration with industry-based patient matching efforts, which will increase buy-in and align with other federal efforts to improve standardization and interoperability.
- Encourages a standardized format for addresses and other data elements.
- Is compatible with existing principles around health data, identity, notice and consent, and interoperability standards, which will ensure consistent deployment across U.S. healthcare organizations.
- Provides guidance on standardization of data capture and best practice processes post mergers, during data conversions, and after closure of an institution.
The Role of Technology
At the organizational level, deployment of the right technologies will go a long way toward resolving patient matching issues. Clearly, a unique identifier on its own cannot adequately address patient misidentification, considering that 95% of organizations responding to the HIMSS/Patient ID Now patient matching survey indicated one was already in place.
What is needed is technology that can catch and correct errors and identify and resolve duplicate and overlaid records which is necessary to improve existing systems, regardless of the presence of a unique patient identifier – something just 20% of survey respondents indicated having. The right technology can also address the privacy and security concerns and ensure data integrity across the continuum.
The goal should be technology that enables end-to-end protection of the master patient index (MPI) and electronic MPI (EMPI) by operating in multiple environments and at multiple stages throughout the patient record process. For example, leveraging biometrics to collect a photo along with the information needed to create a patient record in the MPI/EMPI, and advanced deterministic and probabilistic matching algorithms to analyze and clean patient data before a record is updated. Or leveraging text messaging to send the patient a link to take and submit a selfie and photo of their driver’s license to validate their identity and search for any record matches before assigning biometric credentials to new patients.
In an end-to-end protection model, mismatches are caught upfront, which is something that EHRs cannot do on their own. Most EHR systems feature patient lookup functionality that requires specific processes and data to be precisely entered field by field. If even one detail is off, a search may yield invalid results and can lead to the creation of a new, duplicate patient record or worse, selection of the wrong patient creating an overlay.
Dynamic patient lookup solutions, on the other hand, return instant patient results as they are typed into the system search bar – just like a web browser. This allows everyone involved in the patient matching process to narrow and refine results as they type to achieve positive patient identification.
A Grassroots Solution
The grassroots solution to the nation’s patient identification problem is accurately matching the patient to their medical record at the outset – whether a unique patient identifier is in use or not. The vast majority (90%) of patient record errors begin at registration, making it a critical moment for eliminating medical errors, unnecessary costs, and safety issues associated with a duplicate riddled MPI.
Putting in place the processes, standards, and technologies to enable accurate patient matching at registration prevents downstream contamination that can impact multiple other departments, including clinical, imaging, and billing. It can also enhance revenue cycle efficiencies, reducing days in A/R and decreasing denials. Finally, positive patient identification enables digital transformation across the healthcare system, leading to better care and decreased costs, as well as improved interoperability, patient and provider engagement, and patient access.