Ms. Goodwin is manager of clinical risk and patient safety at Cambridge Health Alliance, Cambridge, MA.
Optimize patient identification by following these steps.
In 2007, the Journal of Urology published an article about patient identification errors among prostate needle core biopsy specimens (J Urol 2007; 178:1245-8). The article identified three patients whose biopsy results were positive for adenocarcinoma, yet upon further treatment-either a radical prostatectomy or radiation therapy-no tumor was found.
Root cause analysis (RCA) on patient one was unable to determine where and when the tissue switch occurred. RCA on patient two determined that the pathologist dictated the patient’s biopsy report while looking at biopsy slides of another patient. RCA on patient number three showed that a pathologist released a report that was not transcribed accurately.
Fast forward a little over a decade to an article in the St. Louis Post-Dispatch describing a lab mix-up in the Midwest that identified a patient with invasive and aggressive prostate cancer on biopsy, but upon radical prostatectomy, no cancer could be found (bit.ly/biopsyerror). The article implies a settlement was reached between the patient and his urologist, the lab, and the hospital with which the urologist was affiliated. The patient, who endured an unnecessary removal of his prostate, continues to struggle with incontinence and erectile dysfunction. Concurrently, there was another man out there who was told he is cancer-free, and actually was not. How can this continue to occur, you may be asking.
Lab and pathology testing may seem like one of those areas where if an error is impacting patient diagnosis or treatment, a urologist might not be on the hook. After all, assuming you take the appropriate specimen from the patient, it is then likely passed off to someone for processing, and perhaps then a courier who is responsible for getting it to the lab. Next, it is in the hands of the lab staff and ultimately a pathologist who is trained in making a diagnosis after analyzing the specimen. The process, of course, is far more complex than this, and involves touches by multiple individuals and pieces of software over a period of time.
Time, source of error may be undeterminable
However, as seen in the RCA on patient one above, it may be undeterminable when or where an error with tissue identification occurred. Lab errors involving patient misidentification and labeling can occur in pre-laboratory, laboratory, and post-laboratory workflows. Events such as incorrect patient registration, reliance on incorrect patient data, failure to properly check patient identifiers, labeling specimens outside the presence of the patients, and erroneous entry of results into an electronic medical record are just a few examples of where lab mix-ups can occur (J Med Biochem 2017; 36:107-12).
Biopsy mix-ups occur rarely-at less than 1%, according to one researcher. But the effects can be catastrophic (bit.ly/biopsyerror). They place providers and institutions at significant risk for litigation, as evidenced by the average of $880 million in wasted medical treatment costs and $698 million in medicolegal expenditures annually (bit.ly/preventingerrors; J Urol 2015; 193:1170-7). An important consideration in why these types of errors continue to occur over decades is that they may not be readily apparent to the operator when they occur.
Accurate patient identification is an important national patient safety goal with both The Joint Commission and the College of American Pathologists. Though some believe a rate of zero error or zero harm is unattainable due to the significant number of human interactions with each discrete specimen, actions can be implemented to prevent these errors.
The Joint Commission directive is clear that containers used for blood and other biologic specimens are to be labeled in the presence of the patient, whether in the hospital, ambulatory, or lab setting. This requires identifying the patient at the time of specimen collection by two identifiers, typically name and date of birth. Notably, a hospital district in Texas has reported having almost 2,500 patients named Maria Garcia and over 9% of these individuals share the same date of birth, posing a greater risk for patient misidentification (J Med Biochem 2017; 36:107-12).
Steps to take
Each organization has different processes, technology, and systems in place for the handling of surgical and other biologic specimens. Use of hospital- or clinic-approved workflows and avoiding workarounds is advised and may include practices such as: use of barcode technology and avoidance of handwriting, confirmation of specimen and associated patient by more than one health care worker and in the presence of the patient, dedicated workspace for processing each specimen, ability for EMR order and information to crosswalk into lab software to avoid transcription errors, color-coded lab cassettes, policy on avoiding distractions at critical times of patient identification, real-time audits and direct observation of processes for immediate feedback and correction of issues, and promotion of active engagement during all processes.
Although the percentage of prostate biopsy mix-ups is low, each one is significant, potentially impacting two patients, not just one. The process of collecting a specimen from a patient and the subsequent handling and processing before a result is ready is rife with human interaction. All organizations should view proper patient identification as a patient safety imperative and work to optimize this process as much as possible.