Eliminating the Damage of Incomplete Data Capture
By Mackenzie Higgins, RHIA, Identity Manager
The lack of standardized data capture policies and processes across information systems has created a patient-matching nightmare for many HIM departments faced with untangling the mess incomplete data capture creates in their EMPIs. While the task may seem hopeless, it’s a great opportunity to suss out the root causes and establish best practices.
Whether driven by a need to optimize data integrity, a looming merger or EHR implementation, or the ONC’s push for healthcare organizations achieve a duplicate rate of less than 1% this year, HIM departments are under pressure to eliminate the flotsam and jetsam polluting their EMPIs. Cleanups, however, can only correct the immediate problem. The long-term solution is to ensure the capture of complete and accurate patient information at the front end – before a record is created.
The two primary problem areas when it comes to incomplete data capture are the patient’s middle name and social security number (SSN). Too often, the middle name is skipped over entirely or just the first initial is recorded, which makes it problematic to distinguish one patient from another with the same first and last name.
When it comes to the SSN, the problem typically lies with the patient’s hesitancy to give out that information. The very real threat of identity theft has made many people reluctant to share their SSN publicly, no matter how advanced the facility’s data security might be. However, without the SSN, it’s difficult to match with confidence the registering patient with an existing medical record.
Lack of standardization is the third challenge. This is commonly considered an IT issue: some systems enable patient record creation with the barest of data, such as a middle initial only, while others have designated some fields and information as required (e.g. the full middle name) before a record can be created. However, the human factor comes into play as well. Rarely is there standardized data capture from one department or facility to the next. Inpatient registration may require the patient’s formal full name, for example, while the affiliated outpatient clinic may only require initials or allow nicknames to be entered.
Unless someone comes along behind scheduling or registration and fills in the demographic gaps created with incomplete data, the resultant shell record will take up permanent residence in the EMPI. There, it wreaks wide-spread havoc that can cost time and money and possibly endanger patients. Most notably, shell records can boost the duplicate rate and increase the volume of false positives tagged by duplication detection algorithms.
The resources required to correct the problems caused by shell records are enormous – problems I witnessed first-hand during one of Just Associates’ recent client engagements. The algorithm we were using was struggling to identify true possible duplicates. Limited required data fields resulted in a significant volume of false positive results, which required us to intervene manually to validate and reconcile the records.
As noted earlier, removing the offending records addresses less than half the problem. Shell records that cannot be validated with confidence remain in the system. Too, without proper procedures and policies in place to ensure complete data capture, new shell records will be created as fast, if not faster, than the existing ones can be removed.
Happily, incomplete data capture is not a foregone conclusion. While not much can be done to speed up data standardization, steps can be taken to improve internal processes. First and foremost, policies and procedures must be established to guide data capture—and they must be communicated to and understood by everyone who potentially is creating patient records.
Special attention should be given to developing strategies for increasing SSN capture. A primary element in nearly all patient matching algorithms, the SSN is crucial for positive patient identification. As such, registration and scheduling staffs should be provided talking points stressing the importance of collecting the SSN to ensure patient safety and the provision of quality care.
Policies should also emphasize the importance of capturing complete data in addition to the SSN. This includes complete middle names vs. initials, as well as correct addresses, phone numbers, hyphenations and even nicknames. Ideally, several forms of identification should be agreed upon to serve as the source of truth to validate any information provided by the patient.
When possible, drill down into duplicate analytics to identify where shell records are coming from most often, and conduct targeted education with that department to ensure everyone is properly trained. Audit regularly for accuracy and completeness and provide more training as needed.
Eliminating the human causes of incomplete data capture through education and training, and enforcing policies and procedures is the best way to ensure the integrity of patient data for the long term.