The Cost of Duplicate Medical Records and Overlays in Healthcare

THE COST OF Duplicate Medical Records and Overlays in Healthcare Duplicate Medical Records Exactly what is a "duplicate medical record?" Quite simply, a duplicate medical record is when a patient is assigned more than one medical record number. They pose a danger to safety if a patient is treated with missing information, (e.g. incorrect blood types, severe allergies), cause a delay in treatment or tests to be repeated, and have a large financial impact on healthcare organizations. Do you know how a duplicate midical record can affect your health? Duplicate records are rampant in healthcare. AVG Hospital 800,000 Total Records 8-12% 64,000 96,000 Duplicates ?. Most hospitals underreport the number of duplicates in their system. Most hospitals underreport the number of duplicates in their system mainly because the algorithms they use aren't powerful or sophisticated enough to identify them. As hospitals share information, duplicates and overlays will grow exponentially. The rise of Integrated Delivery Networks (IDNS) and Health Information Exchanges (HIES) is exponentially increasing the size of enterprise master patient indexes (EMPIS) from disparate sources causing more overlays and duplicates. Cultural naming conventions can be a driving force for the creation of duplicate medical records. In a 2011 study, Harris County, TX researched a number of years of data and determined that they had over 2,400 patients named Maria Garcia and of those, 231 had the same birth date. 92% of duplicates are created during inpatient registration patient A study was conducted by Johns Hopkins Hospital to investigate the role that registration process plays in the creation of duplicate records. They found that 92% of the errors resulting in duplicates over the course of the fiscal year occurred during inpatient registration. Overlays Widely considered the most troublesome patient identity problem, "overlays" in healthcare occur when two different patient records are erroneously identified as one individual. This can be particularly dangerous when treating a patient based on the wrong medical record. There is a higher risk for duplicates and overlays to be created at pediatric hospitals. Children's hospitals are most susceptible to duplicates and overlays largely because children lack official forms of identification. Their caregiver may not know accurate demographic data of the child, and a child doesn't have a formal mode of identification. Incorrect dates of birth are 10% more likely to occur on pediatric records than on adult records. Children can't often speak for themselves. This can lead to incorrect medication and duplicate diagnostic tests. The implications of duplicate medical records are widespread. Patients are often mistreated with missing or incorrect information which can be very dangerous. 3 Wrong blood type Patients could be mistreated due to the incorrect blood type documented on their electronic medical record. Allergies Medications could be prescribed that are dangerous to a patient with allergies, jeopardizing safety. a X-ray exposure Incomplete records could result in repeat tests that are costly and potentially dangerous. Drains available resources Repeat tests and procedures decreases available staff and resources to treat other patients. Unintended injury or illness Nearly 1/5 of CIOs surveyed in a recent poll say they can attribute at least one adverse event to a patient mismactch within the last year. The Cost Overlay and duplicate medical records can lead to reimbursement losses, administrative inefficiencies, resource drains, liability concerns, and perhaps most significantly, compromised care delivery and threats to patient safety. The biggest threat to the creation of duplicates and overlays in healthcare is when a hospital adopts an electronic health record system (EHR). » $50 According to Fox and Sheridan, the average cost of a duplicate medical record pair is $50. If the records aren't reconciled, the costs are even higher. Time lost by a patient and provider is often overlooked as a consequence of duplicate medical records. Correcting a paper-based overlay patient record can take 60- 100 hours. 60 100 hours In an EHR environment, the time to fix an overlay can take months depending on the complexity of the system. months In a recent survey by The College of Healthcare Information Management Executives (CHIME), respondents indicated that although "data cleansing" is a marginal component of other duties, they have 2 or more dedicated personnel. 2 or more dedicated personnel A recent overlay case involving twin girls from the Children's Medical Center of Dallas took 16 staff members working 3 months to correct the problem. 16 staff members working 3 months A study conducted at Children's Medical Center in Dallas found that cost reflected in patient records associated with repeated tests or treatment delays were, on average, $1099 each with one repeated tests in ten associated with bad debt. $1099 each with one in ten associated with bad debt. What often goes unmeasured is the amount of trust that can be lost from the creation of duplicates and overlays. Physicians and other clinicians need to trust the information they access from healthcare information systems - dulplicates and overlays undermine that trust. The Solution Despite the fact that duplicates, overlays, and overlaps are known to be a pressing problem in the healthcare industry, we still continue to see unintended injuries and illnesses resulting from patient data-matching errors. In the absence of patient identification industry standards or national leadership on mandating unique patient identifiers, what current technology options are used by healthcare facilities for patient data matching? Patient Smart Cards Algorithms Biometrics The College of Healthcare Information Management Executives (CHIME) recently conducted a survey seeking information on the current state of patient data matching methodologies. 64.8% 68% of hospitals currently use some kind of unique patient identifier to match patient data. 50.8% 50.8% of the respondents indicated that they use probabilistic algorithms to match patient data. 34.4% 34.4% use deterministic matching 5.5% 5.5% use biometric matching strategies Algorithms There are three common types of algorithms used in hospital information system (HIS) software: deterministic, rule-based, and probablilistic matching. Probabilistic Algorithms 95% accuracy Probabilistic algorithms generally provide accuracy rates of up to 95% or higher. These algorithms determine precise record linkages by using complex mathematical principles to help analyze organization-specific data. They incorporate the unique characteristics of each database into the matching process using automated, statistically valid and highly accurate processes, rather than relying on more subjective and error prone manually configured matching rules and "confidence levels". Deterministic Algorithms Deterministic algorithms are the most common, but generally yield about a 50% to 60% accuracy rate because they require exact or phonetic matches on certain data elements. 50% to 60% accuracy Rules-based Algorithms 70% 80% accuracy to Rules-based algorithms offer a high accuracy rate - typically about 70% to 80% – because of a more advanced matching method that utilizes pre-set confidence levels for certain data elements. Smart Cards Smart cards look very similar in size and shape to credit cards and include an embedded, integrated circuit chip that can be either a microcontroller chip with internal memory or a secured memory chip alone. The card communicates with a reader either through direct physical contact or with a remote contactless electromagnetic field that energizes the chip and transfers data between the card and the reader. Smart cards are widely acknowledged as one of the most secure and reliable forms of an electronic identification token. A smart card can combine serveral ID technologies, including the embedded chip, visual security markings, magnetic strip, barcode, and/or an optical strip. Smart cards can authenticate identity by storing encrypted personal demographic information such as date of birth, sex, and race. Smart cards are an effective, secure means of identification, but what happens if it's lost, stolen, or you forget to bring it with you to the hospital or doctor's office? Biometrics Biometrics are increasingly used in healthcare as a method to identify patients. Physiological characteristics of the human body for identification (e.g. fingerprint, palm vein, or iris), can be used by healthcare facilities to seamlessly identify a patient by scanning their biometric identity. In some cases, they can automatically pull up the patient's electronic medical record after verifying who they are. Non-contact iris biometrics for patient identification supports hospital infection control initiatives plus it utilizes 1:N biometric matching which is the only true way to prevent duplicates, overlays, and healthcare fraud at the point of service. Iris biometrics also has a low occurrence of false positives and extremely low, almost 0% false negative rate. Iris Palm vein biometric identification offers a viable solution for patient identification in healthcare, but relies on 1:Few matching technology which can't completely prevent duplicate medical records, overlays, or healthcare fraud at the point of service. Palm Vein Finger Print Fingerprints are the most well-known biometric modality, but require physical contact with a hardware device which is not conducive to infection control in a hospital setting. There are three types of biometric matching types. 1:1 Verification 1:1 (one-to-one) verification cannot completely prevent duplicate medical records, overlays, or patient fraud because it is only ever comparing one scanned template against one enrollment (stored) template. This biometric matching type answers the question, "Are you who you say you are?" 1:Few Segmented Identification In a 1:Few segmented identification system the entire biometric database is not searched. For this reason, a patient could enroll the same biometric data multiple times under different credentials and the system would never recognize duplicates. This biometric matching type answers the question: "Are you in this group?" 1:N Identification 1:N (one-to-many) identification compares the captured biometric template against all stored templates. No other information is required besides the biometric scan and it is the only true way to prevent duplicates, overlays, and healthcare fraud at the point of service. This biometric matching type answers the question, "Who are you?" Sources "What You Don't Know Can't Hurt You", Executive Insight 01/17/12 http://healthcare-executive-insight.advanceweb.com/Features/ Articles/What-You-Dont-Know-Can-Hurt-You.aspx Are you who you claim to be? http://campus.ahima.org/audio/2009/RB)72109.pdf Smart Card Alliance, http://www.smartcardalliance.org/pages/smart-cards- faq#why-are-smart-cards-better-than-other-id-token-technologies "Do You Know Who's In Your EHR?", Just Associates http: /kamfE M2SYS Healthcare Solutions podcast with Beth Just from Just Associates Duplicate Medical Records: A Survey of Twin Cities Healthcare Organizations http://www.ncbi.nim.nih.gov/pmc/articles/PMC2815491 Fierce Health IT, http://www.fiercehealthit.com/story/which-maria-garcia- bipartisan-center-seeks-improve-patient-data-matching/2012-06-27 CHIME Survey on Patient Data Matching, 05/16/12 http://www.cio-chime.org ladvocacy/resources/download/Summary_of CHIME Survey on_Patient Data.pdf AHIMA.org Brought to you by: M2SYS Healthcare Solutions and M2SYS RIGHTPATIENTTM Multi-modal healthcare biometrics for patient identification. You Tube RightPatient GRightPatient www.youtube.com/user/M25YS ..... ... ...... ...