Mind the Gap

Electronic health records have the potential to transform medicine by preventing errors, facilitating current best practices, and enabling research that will determine future evidence-based practices. If the data are to be usable for these purposes, however, they must be present and accurate. In this issue, Ehrenfeld et al.1 contribute a multicenter investigation that adds to our knowledge regarding the incidence and underlying causes of gaps in anesthesia information management system (AIMS) documentation. This investigation confirms the work of their own group and others demonstrating gaps between guidelines and clinical practice,2 between anesthesia record documentation and veracity,3,4 and in the nascent efforts to implement AIMS-based decision support.5–7 The publication of this paper is also timely. Without urgent action by anesthesiologists and the AIMS vendors, these systems will not reach their full potential. If anesthesiologists were to be excluded from financial incentives for implementation of AIMS, this would perpetuate the low rate of AIMS adoption that currently exists.8,9 Hence, the work of decades of AIMS research will fail to serve the best interests of our patients and specialty. The authors report a high incidence of gaps in arterial blood pressure (BP) documentation of at least 10 minutes during anesthesia maintenance, with the rate varying among institutions and among practitioners within institutions. The incidence of shorter gaps (5 to 10 minutes in duration) and of those occurring prior to anesthetic induction are likely much higher. While technically consistent with American Society of Anesthesiologists monitoring standards that allow brief interruptions in BP monitoring, the authors' secondary analysis demonstrates that many of the observed gaps were potentially avoidable or lacked supporting documentation. Many gaps were therefore deviations from existing monitoring standards that increase liability exposure for practitioners and institutions.10,11 The authors are to be commended for a scientific analysis of this issue. Some monitoring gaps are unavoidable because of limitations in technology,2 and others are clearly inappropriate and require corrective action. For example, patients awaiting a postanesthesia care unit bed following completion of surgery should be monitored at least per postanesthesia care unit standards, or documented in a compliant manner if anesthesia care is being continued. Other monitoring gaps are explicable, but perhaps longer than clinically necessary. Does prone patient positioning and padding need to be absolutely complete before resuming BP measurement? Could BP measurement continue on another extremity while intra-arterial pressure monitoring is established? Some monitoring gaps are avoidable entirely. BP measured, but not recorded because of AIMS hardware/software failure, can usually be retrieved from hemodynamic monitors and manually transcribed into the AIMS. Perhaps the most striking finding by Ehrenfeld et al. was not in the AIMS records, but in the sample of intercurrent handwritten records in which not a single BP was missing—a credibility gap. The evidence from previous investigations is clear on this point,3,12–17 and is an indictment (if not a conviction) of 110 years of paper anesthesia records filled with “interpolated” and “smoothed” physiologic data that are largely fictional. It would be unfair, however, to single out anesthesia professionals in this regard. One frequently sees history and physical examination notes by practitioners who carry neither stethoscope nor penlight documenting “lungs clear to auscultation” and “PERRLA” (pupils equal, round, reactive to light and accommodation). Ehrenfeld et al. add to the evidence that the expectation of uninterrupted anesthesia physiologic data documentation may be unreasonable, and that smooth records are often unbelievable. Regulatory agencies, payers, professional standards groups, and the legal tort system must change accordingly if we expect health care professionals to alter their “defensive” behaviors. Many anesthesiologists hesitate to adopt AIMS because of medicolegal concerns, despite millions of anesthetics documented by AIMS without detrimental malpractice effects. There is also an AIMS technology implementation gap. Why is there so little AIMS-based decision support? Clinical alerts and other forms of decision support are likely the most significant benefits of using an AIMS. AIMS-based decision support should be the nexus of “smart alarms” that minimize the frequency of false alerts that lead to inattention. Previous reports indicated that AIMS-based alerts improved clinical and administrative processes, and led to sustained behavioral changes.5–7,18–20 In the current report, the authors implemented alerts that reduced gaps in BP documentation. Several years and several software upgrades after many of the earlier reports, there is still no industry standard that AIMS offer such features. AIMS (and other electronic health record) vendors in the United States (U.S.) are not regulated for safety by the Food and Drug Administration; they are only required to adhere to quality control and manufacturing practices that focus on the development process.a With little external pressure from regulators or anesthesiologists, many AIMS vendors have not implemented decision support capabilities, despite evidence of their benefits. Why are so few AIMS installed worldwide? AIMS installation expenses are a major barrier to adoption.8 Moreover, there is great doubt that U.S. anesthesiologists will qualify as eligible practitioners and that AIMS will be certified products for U.S. monetary incentives for electronic health record implementation under the HITECH act.b An urgent cooperative effort between anesthesiologists and industry will be necessary to advocate for these funds. Anesthesiologists must insist that “meaningful use” criteria include measures that are relevant to perioperative care, and AIMS vendors must integrate practical decision support tools for anesthesiologists to implement. Anesthesiologists were among the earliest adopters of electronic health records and continue to lead in conducting research that demonstrates their merits. It would be ironic if we were to be excluded from incentives for AIMS adoption and were to fall behind the rest of the health care profession. Mind the gaps. DISCLOSURES Name: David L. Reich, MD. Contribution: This author helped write this editorial. Attestation: David L. Reich approved the final manuscript. Name: David B. Wax, MD. Contribution: This author helped write this editorial. Attestation: David B. Wax approved the final manuscript.