In Response

    loading  Checking for direct PDF access through Ovid

Excerpt

Thank you for taking the time to carefully evaluate our publication. Electronic medical record (EMR) systems and anesthesia information and monitoring systems have held out the promise of streamlining data collection for various registries for years. Unfortunately, to date, this promise has been far greater than reality. Collecting and transmitting enormous quantities of data to collection centers such as the Multicenter Perioperative Outcomes Group and the Anesthesia Quality Institute depends in large part upon marrying administrative data such as diagnoses and procedures with clinical data. Multiple investigations have shown that this frequently results in inaccurate information being assessed because of deficiencies in the coding process at the hospital level and in inherent weaknesses in the codes themselves. Pediatric cardiac disease is one of the most complex to code because of the myriad of diagnoses and procedures. Efforts to standardize the nomenclature through the International Paediatric Cardiac and Congenital Cardiac Code have been very helpful, but this standard has not yet been fully implemented in the International Classification of Diseases (ICD) coding systems that efforts such as Multicenter Perioperative Outcomes Group and the Anesthesia Quality Institute rely upon, including the recent ICD-10 release.1
In addition, the harvest process fails to adequately incorporate an agreed upon adverse event terminology, and many hospital systems are reluctant to release this protected information even in a blinded fashion. What these data collection agencies can excel at is the amassing of enormous amounts of numeric data elements such as vital signs, drug usage, and dosage (including inhaled anesthesia concentrations) and demographic elements such as age at surgery and sex, lengths of stay, and certain very specific care elements that are documented discretely.2 The recent publication “Reference Values for Noninvasive Blood Pressure in Children During Anesthesia: A Multicentered Retrospective Observational Cohort Study” by de Graaff et al3 is an excellent example of using “big data” harvesting to analyze these discrete data without the bias of relying upon human-based administrative procedural coding that most frequently occurs post hoc by nonclinical personnel abstracting information from clinical records for the purposes of hospital billing.
Another element you raise is the sharing of data and outcomes across multiple institutions. Assuming hospital systems have a business agreement and are on the same EMR platform (such as EPIC), it is sometimes possible for there to be a seamless transmission of data from one institution to another, but in practice, this has proved to be less than ideal and to date there is no way for competing EMR platforms such as EPIC and Cerner to communicate effectively. In addition, there has been a reluctance on the part of EMR platforms and some medical registries to directly communicate data because of concerns about data integrity and accuracy in the medical records being transmitted without any sort of auditing of the data for clinical accuracy.
The authors wholeheartedly agree with your suggestion of revisiting the “Common Rule” regulations to allow for better long-term outcome assessment, particularly in a field such as congenital cardiac disease where outcomes are properly measured across decades and multiple institutions.
    loading  Loading Related Articles