Letters to the Editor: Letters & Announcements
To the Editor:
I read with interest the article by Dr. Feldman (1). Despite the author’s effort to address an important topic, the approach to answering this complex question has resulted in a potentially misleading conclusion. The article indicates that departments using AIS find that it does not interfere with “risk management.” In addition, in the opinion of some of the surveyed departments, electronic records have been helpful in disposing of potential and actual malpractice actions more favorably. Unfortunately, there are substantial limitations in the study methodology that preclude definite conclusions. This is due in part to many currently unresolved issues associated with the intermeshing of medical informatics and anesthesiology. A number of these issues are enumerated below.
First, there is still much to learn about managing biosignals transduced to a real-time monitor, let alone a written record with stored data. For example, at this time all biosignals are analog in nature. These signals must be electronically converted to digital signals to be recorded by a PC (or any computer) (2). In addition there are real-time data sampling and data quantification problems that can significantly alter read-out values (2). Consequently, data printed on a written record from stored computer data is, by its intrinsic nature, data that has been manipulated into a form that may or may not be sufficiently accurate to reflect the clinical situation in real time. This problem of computer data manipulation is a limitation that is similar to the current hand written record. It is not possible to know from the information presented whether the AIS surveyed are actually more accurate than a written record.
Second, this paper does not define the term “risk management.” Risk management, as a field of study, incorporates specific methods for identifying and managing issues in an attempt to decrease not only the risk of lawsuits but also for identifying and mitigating many other types of liability issues (3). In any JCAHO accredited institution, risk management methods are regularly applied to problems in the medical setting not related to provider negligence. Perhaps substituting the words “decrease in liability” or “decreasing liability” for “risk management,” in this article, would produce a more understandable (but not more valid) interpretation of the survey data. Absent a specific definition of “risk management” in the questionnaire, the respondents are left to apply their own interpretation to a term with several contextual definitions, which profoundly confounds our ability to interpret the survey results.
Third, the survey asks the questionnaire respondents to come to a legal conclusion about whether the increase in quality of documentation (which is presumed but not established by objective evidence) made a difference in legal actions such as voluntarily dismissing a suit or settling a suit by somehow decreasing liability. It seems exceedingly unlikely that the respondents to this survey could definitively know the plaintiff’s actual reasoning for dropping or settling a suit and whether it was actually related to AIS, since most physicians do not understand the potential legal proof problems associated with these electronic records. The AIS are proprietary and they vary significantly because there is no established standard for these systems. More importantly, using AIS is not the standard of care currently. A well-qualified attorney could (and likely would) raise substantial questions about the accuracy and the validity of any (AIS) system based on the aforementioned factors. Therefore, depending on the relationship between the anesthesia record and the particular liability issue in question, the use of AIS could influence the decision of opposing counsel in either direction: to pursue or reject a claim, depending on the nature of the injury sustained and the ability to prove the relationship (in legal terms, the attachment of liability) between the presumed negligence and the injury (proximate cause).
Fourth, this paper speculates incorrectly that any additional data not printed on the anesthesia record may, in the future, be introduced into a legal proceeding in a manner that might be determined by the institution or department in question. To suggest that a set of circumstances would arise where part of the record could be voluntarily withheld is incorrect. Any additional data archived in an electronic database of the AIS is now and will in the future be legally available for analysis in any litigation. A patient’s record and all data contained therein belong to the patient (4). The complete record cannot be sequestered or remain undisclosed without creating a situation where the institution and/or the physician are in contempt of the court.
The methodology of this paper severely constrains our ability to derive meaningful conclusions about AIS and its role in liability reduction and risk management. Additional investigation that addresses the issues noted above will be required before we can make a valid determination of the utility of AIS for liability reduction and risk management.
Philip Edward Lane, MD, JD, MPH, MBA
Department of Anesthesiology; Rush University Medical Center; Chicago, IL; email@example.com
1. Feldman JM. Do anesthesia information systems (AIS) increase malpractice exposure? Results of a survey. Anesth Analg 2004;99:840–3.
2. Van Bemmel JH, Musen MA. Eds. Biosignal analysis in handbook of medical informatics. New York: Springer, 1997:120–5.
3. Vincent C. The development of clinical risk management. In: Clinical risk management. London: BMJ Books, 2001:45–60.